Merge remote-tracking branch 'origin/main' into brennanb2025/reopen-closed-tab

# Conflicts:
#	src/renderer/src/store/index.ts
#	src/renderer/src/store/slices/diffComments.test.ts
#	src/renderer/src/store/slices/store-test-helpers.ts
#	src/renderer/src/store/slices/terminals.ts
#	src/renderer/src/store/types.ts

.github/workflows/computer-e2e.yml

@@ -10,7 +10,14 @@ on:
       - 'config/scripts/computer-use-skill-guidance.test.mjs'
       - 'config/scripts/computer-use-smoke.mjs'
       - 'config/scripts/computer-use-smoke.test.mjs'
+      - 'config/scripts/daemon-boot-smoke.mjs'
       - 'config/scripts/verify-computer-native.mjs'
+      # Why: the native-smoke job boots the built terminal daemon under plain
+      # Node, so any change to the daemon bundle graph or the main build must
+      # re-run it (the v1.4.129-rc.1 daemon outage shipped with green CI).
+      - 'electron.vite.config.ts'
+      - 'build-plugins/**'
+      - 'src/main/daemon/**'
       - 'native/computer-use-macos/**'
       - 'native/computer-use-linux/**'
       - 'native/computer-use-windows/**'
@@ -98,6 +105,12 @@ jobs:
       - run: pnpm verify:computer-native
       - run: pnpm build:cli
       - run: pnpm build:electron-vite
+      # Why: boot the BUILT daemon-entry under plain Node the way production
+      # forks it. v1.4.129-rc.1 shipped a daemon that exited at module load
+      # (leaked electron require) while every other check passed; this fails
+      # the PR when the built daemon cannot start on ubuntu-22.04 / windows.
+      - name: Daemon boot smoke
+        run: node config/scripts/daemon-boot-smoke.mjs
       - if: runner.os == 'Linux'
         env:
           ORCA_COMPUTER_E2E: '1'

.github/workflows/daemon-relocation-spike.yml

@@ -0,0 +1,97 @@
+name: Daemon Relocation Spike
+
+# Why: Phase 1 of the Windows update-survival work relocates the terminal
+# daemon's file closure out of the install dir so it survives an update. Before
+# implementing it in the app, this spike finds the MINIMAL set of packaged files
+# that a copied Orca.exe (run as node) needs to start the daemon and drive a
+# ConPTY session from a relocated directory, holding no install-dir file locks.
+# Runs from the feature branch via push (workflow_dispatch only works on the
+# default branch); main is never touched.
+
+on:
+  push:
+    branches:
+      - Jinwoo-H/windows-update-survival
+    paths:
+      - 'tools/daemon-relocation-spike/**'
+      - '.github/workflows/daemon-relocation-spike.yml'
+  workflow_dispatch: {}
+
+permissions:
+  contents: read
+
+concurrency:
+  group: daemon-relocation-spike-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  spike:
+    name: relocation file-set spike
+    runs-on: windows-2022
+    timeout-minutes: 40
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+        with:
+          # This job only reads the repo to build/run the spike; it never pushes.
+          persist-credentials: false
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version-file: package.json
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v6
+        with:
+          run_install: false
+
+      - name: Install dependencies
+        run: pnpm install --frozen-lockfile
+
+      # Why: cache the unpacked build keyed on the inputs that affect it, so
+      # spike-script-only edits skip the ~15 min electron-builder build.
+      - name: Cache unpacked build
+        id: cache-unpacked
+        uses: actions/cache@v4
+        with:
+          path: dist/win-unpacked
+          key: win-unpacked-${{ hashFiles('src/**', 'config/**', 'package.json', 'pnpm-lock.yaml') }}
+
+      - name: Build unpacked app
+        if: steps.cache-unpacked.outputs.cache-hit != 'true'
+        run: pnpm run build:unpack
+
+      - name: Run relocation spike (all tiers)
+        shell: pwsh
+        run: |
+          New-Item -ItemType Directory -Force artifacts | Out-Null
+          $log = "artifacts/spike-output.log"
+          $tiers = @('full', 'no-gpu', 'minimal')
+          $any = $false
+          foreach ($tier in $tiers) {
+            "===== TIER: $tier =====" | Tee-Object -FilePath $log -Append
+            $work = Join-Path $env:RUNNER_TEMP "spike-$tier"
+            # --keep-work-dir so the per-tier daemon stdout/stderr logs survive
+            # for the artifact upload (the spike otherwise removes work-dir),
+            # which is what diagnoses a tier that fails to reach ready.
+            node tools/daemon-relocation-spike/spike.mjs `
+              --app-dir dist/win-unpacked `
+              --work-dir "$work" `
+              --tier $tier `
+              --keep-work-dir 2>&1 | Tee-Object -FilePath $log -Append
+            if ($LASTEXITCODE -eq 0) { $any = $true }
+          }
+          if (-not $any) { throw "No tier passed the relocation spike" }
+
+      - name: Upload spike output
+        if: always()
+        uses: actions/upload-artifact@v7
+        with:
+          name: daemon-relocation-spike-output
+          path: |
+            artifacts/spike-output.log
+            ${{ runner.temp }}/spike-*/**/*.log
+          retention-days: 7
+          if-no-files-found: warn

.github/workflows/mobile-android-release.yml

@@ -9,22 +9,8 @@ on:
       - 'mobile-android-v*'
   workflow_dispatch:
     inputs:
-      bump_patch_version:
-        description: 'Bump the mobile marketing version patch number before release'
-        required: false
-        default: false
-        type: boolean
       release_version:
-        description: 'Optional exact mobile marketing version override, e.g. 0.0.15'
-        required: false
-        type: string
-      bump_android_version_code:
-        description: 'Increment the Android versionCode before release'
-        required: false
-        default: false
-        type: boolean
-      android_version_code:
-        description: 'Optional exact Android versionCode override'
+        description: 'Optional exact mobile marketing version assertion, e.g. 0.0.22'
         required: false
         type: string
       publish_github_release:
@@ -69,9 +55,6 @@ jobs:
         id: release
         env:
           MOBILE_ANDROID_RELEASE_VERSION: ${{ github.event.inputs.release_version }}
-          MOBILE_ANDROID_BUMP_PATCH_VERSION: ${{ github.event.inputs.bump_patch_version }}
-          MOBILE_ANDROID_VERSION_CODE: ${{ github.event.inputs.android_version_code }}
-          MOBILE_ANDROID_BUMP_VERSION_CODE: ${{ github.event.inputs.bump_android_version_code }}
           MOBILE_ANDROID_PUBLISH_RELEASE: ${{ github.event.inputs.publish_github_release }}
         run: node scripts/prepare-android-release.mjs
 

.github/workflows/pr.yml

@@ -66,6 +66,17 @@ jobs:
       - name: Check styled scrollbars
         run: pnpm check:styled-scrollbars
 
+      - name: Check reliability gate manifest
+        run: pnpm run check:reliability-gates
+
+      # Why: oxlint fails any file over max-lines that is NOT suppressed, so this
+      # ratchet forbids ADDING a new suppression (inline disable or mobile max
+      # bump). Existing oversized files are grandfathered in
+      # config/max-lines-baseline.txt, which may only shrink — new bypasses fail
+      # here with a clear message instead of silently growing the debt.
+      - name: Enforce max-lines ratchet (no new bypasses)
+        run: pnpm run check:max-lines-ratchet
+
       # Why: project-owned type declarations must live in .ts so tsc
       # actually checks them. TypeScript's skipLibCheck: true (inherited
       # from @electron-toolkit/tsconfig) silently widens unresolved names

.github/workflows/release-cut.yml

@@ -719,6 +719,10 @@ jobs:
               ~/.cache/electron-builder
 
     runs-on: ${{ matrix.os }}
+    # Why: hosted runners hard-cap jobs at 6h; the Windows SignPath waits
+    # (1h inner + 4h installer) are budgeted to fit under this with the
+    # build itself, so a slow approval can't kill the job mid-flow.
+    timeout-minutes: 360
 
     permissions:
       actions: read
@@ -883,12 +887,32 @@ jobs:
         if: matrix.platform == 'win'
         shell: pwsh
         run: |
-          # Why: Some hosted Windows images start without PSGallery registered.
-          if ($null -eq (Get-PSRepository -Name PSGallery -ErrorAction SilentlyContinue)) {
-            Register-PSRepository -Default -InstallationPolicy Trusted
-          }
+          $ErrorActionPreference = 'Stop'
+          # Why: force TLS 1.2 so gallery downloads work on older hosted images.
+          [Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12
 
-          Set-PSRepository -Name PSGallery -InstallationPolicy Trusted
+          # Why: on some hosted Windows images `Register-PSRepository -Default`
+          # fails inside the legacy nuget.exe provider with "Missing option value
+          # for: '-source'", so PSGallery is never registered and the install
+          # below dies with "No repository with the name 'PSGallery'". PSResourceGet
+          # (bundled with PowerShell 7.4+) has PSGallery registered by default and
+          # avoids that code path, so prefer it and fall back to PowerShellGet only
+          # when it is absent.
+          $useResourceGet = $null -ne (Get-Command -Name Install-PSResource -ErrorAction SilentlyContinue)
+
+          if ($useResourceGet) {
+            if ($null -eq (Get-PSResourceRepository -Name PSGallery -ErrorAction SilentlyContinue)) {
+              Register-PSResourceRepository -PSGallery -Trusted
+            } else {
+              Set-PSResourceRepository -Name PSGallery -Trusted
+            }
+          } else {
+            Install-PackageProvider -Name NuGet -MinimumVersion 2.8.5.201 -Force | Out-Null
+            if ($null -eq (Get-PSRepository -Name PSGallery -ErrorAction SilentlyContinue)) {
+              Register-PSRepository -Default -InstallationPolicy Trusted
+            }
+            Set-PSRepository -Name PSGallery -InstallationPolicy Trusted
+          }
 
           $trimChars = [char[]]@([System.IO.Path]::DirectorySeparatorChar, [System.IO.Path]::AltDirectorySeparatorChar)
           $documentsRoot = [System.IO.Path]::GetFullPath([Environment]::GetFolderPath('MyDocuments')).TrimEnd($trimChars)
@@ -917,7 +941,11 @@ jobs:
             }
 
             try {
-              Install-Module -Name SignPath -Repository PSGallery -MinimumVersion 4.0.0 -MaximumVersion 4.999.999 -Scope CurrentUser -Force -AllowClobber -ErrorAction Stop
+              if ($useResourceGet) {
+                Install-PSResource -Name SignPath -Version '[4.0.0,5.0.0)' -Repository PSGallery -Scope CurrentUser -TrustRepository -Reinstall -ErrorAction Stop
+              } else {
+                Install-Module -Name SignPath -Repository PSGallery -MinimumVersion 4.0.0 -MaximumVersion 4.999.999 -Scope CurrentUser -Force -AllowClobber -ErrorAction Stop
+              }
               Import-Module SignPath -ErrorAction Stop
               Get-Command -Name Get-SignedArtifact -Module SignPath -ErrorAction Stop
               break
@@ -934,6 +962,244 @@ jobs:
             }
           }
 
+      # ── Windows inner-binary signing (issue #7785) ─────────────────────
+      # Why: SignPath cannot deep-sign inside NSIS installers, so inner PE
+      # files (Orca.exe, node-pty *.node, DLLs) are signed via a separate zip
+      # request, then the installer is rebuilt from the signed tree before the
+      # existing installer signing request below. Every step in this chain is
+      # fail-open (continue-on-error + outcome gating): any failure ships the
+      # original installer with unsigned inner binaries, exactly like releases
+      # did before this chain existed. Rehearsed end to end in run 28988432001
+      # (.github/workflows/windows-signing-rehearsal.yml).
+
+      # Why: only unsigned PE files go to SignPath. Files that already carry a
+      # valid signature (Microsoft's OpenConsole.exe) must keep their signer.
+      - name: Stage unsigned inner PE files for signing
+        id: stage-inner
+        if: matrix.platform == 'win'
+        continue-on-error: true
+        shell: pwsh
+        run: |
+          $root = Resolve-Path 'dist/win-unpacked'
+          $stage = New-Item -ItemType Directory -Force -Path 'signing-stage'
+          $list = New-Object System.Collections.Generic.List[string]
+          $skipped = New-Object System.Collections.Generic.List[string]
+
+          Get-ChildItem -Path $root -Recurse -File |
+            Where-Object { $_.Extension -in '.exe', '.dll', '.node' } |
+            ForEach-Object {
+              $relative = [System.IO.Path]::GetRelativePath($root, $_.FullName)
+              $signature = Get-AuthenticodeSignature -FilePath $_.FullName
+              if ($signature.Status -eq 'Valid') {
+                $skipped.Add("$relative  <already signed: $($signature.SignerCertificate.Subject)>")
+                return
+              }
+              $destination = Join-Path $stage.FullName $relative
+              New-Item -ItemType Directory -Force -Path (Split-Path $destination) | Out-Null
+              Copy-Item -Path $_.FullName -Destination $destination -Force
+              $list.Add($relative)
+            }
+
+          if (-not ($list -contains 'Orca.exe')) {
+            throw 'Orca.exe was not staged for signing; unpacked layout changed?'
+          }
+          if (-not ($list | Where-Object { $_ -like '*conpty_console_list.node' })) {
+            throw 'node-pty conpty_console_list.node was not staged; this is the file from issue #7785.'
+          }
+
+          Set-Content -Path 'inner-signing-list.txt' -Value ($list -join "`n")
+          Write-Host "Staged $($list.Count) unsigned PE files for signing:"
+          $list | ForEach-Object { Write-Host "  $_" }
+          Write-Host "Skipped $($skipped.Count) already-signed files:"
+          $skipped | ForEach-Object { Write-Host "  $_" }
+
+      - name: Upload unsigned inner binaries for SignPath
+        id: upload-unsigned-inner
+        if: matrix.platform == 'win' && steps.stage-inner.outcome == 'success'
+        continue-on-error: true
+        uses: actions/upload-artifact@v7
+        with:
+          name: orca-windows-inner-unsigned-${{ needs.cut.outputs.tag }}
+          path: signing-stage/**
+          if-no-files-found: error
+
+      - name: Submit inner binaries signing request
+        id: submit-inner-signing
+        if: matrix.platform == 'win' && steps.upload-unsigned-inner.outcome == 'success'
+        continue-on-error: true
+        uses: signpath/github-action-submit-signing-request@v2
+        with:
+          api-token: ${{ secrets.SIGNPATH_API_TOKEN }}
+          organization-id: c37aa192-a27a-4377-9c90-5d6c95912dc0
+          project-slug: orca
+          signing-policy-slug: release-signing
+          artifact-configuration-slug: windows-inner-binaries-zip
+          github-artifact-id: ${{ steps.upload-unsigned-inner.outputs.artifact-id }}
+          wait-for-completion: false
+
+      - name: Notify Slack that inner-binary signing is waiting for approval
+        id: notify-inner-signing
+        if: matrix.platform == 'win' && steps.submit-inner-signing.outcome == 'success'
+        continue-on-error: true
+        shell: pwsh
+        env:
+          SLACK_WEBHOOK_URL: ${{ secrets.SLACK_WEBHOOK_URL }}
+          SIGNPATH_ORGANIZATION_ID: c37aa192-a27a-4377-9c90-5d6c95912dc0
+          SIGNPATH_REQUEST_ID: ${{ steps.submit-inner-signing.outputs.signing-request-id }}
+          SIGNPATH_REQUEST_URL: ${{ steps.submit-inner-signing.outputs.signing-request-web-url }}
+          TAG: ${{ needs.cut.outputs.tag }}
+          GITHUB_RUN_URL: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}
+        run: |
+          if ([string]::IsNullOrWhiteSpace($env:SLACK_WEBHOOK_URL)) {
+            throw 'SLACK_WEBHOOK_URL secret is required so release approvers know when SignPath is waiting.'
+          }
+
+          $requestUrl = $env:SIGNPATH_REQUEST_URL
+          if ([string]::IsNullOrWhiteSpace($requestUrl)) {
+            $requestUrl = "https://app.signpath.io/Web/$env:SIGNPATH_ORGANIZATION_ID/SigningRequests/$env:SIGNPATH_REQUEST_ID"
+          }
+
+          $message = "Orca Windows release $env:TAG inner-binaries signing request (1 of 2) is ready for SignPath approval.`n<$requestUrl|Open SignPath signing request>`n<$env:GITHUB_RUN_URL|Open GitHub Actions run>"
+          $payload = @{
+            text = $message
+            blocks = @(
+              @{
+                type = 'section'
+                text = @{
+                  type = 'mrkdwn'
+                  text = $message
+                }
+              }
+            )
+          } | ConvertTo-Json -Depth 5
+
+          Invoke-RestMethod -Method Post -Uri $env:SLACK_WEBHOOK_URL -ContentType 'application/json' -Body $payload
+
+      # Why gate on the notify outcome too: if nobody was told to approve,
+      # don't hold the release for the approval window — fall through and
+      # ship like today instead. The 1h wait (vs the installer's 4h) keeps
+      # both waits plus the build inside the 360-minute job cap; missing it
+      # falls through to today's unsigned-inner flow rather than blocking.
+      - name: Download signed inner binaries from SignPath
+        id: download-signed-inner
+        if: matrix.platform == 'win' && steps.submit-inner-signing.outcome == 'success' && steps.notify-inner-signing.outcome == 'success'
+        continue-on-error: true
+        shell: pwsh
+        env:
+          SIGNPATH_API_TOKEN: ${{ secrets.SIGNPATH_API_TOKEN }}
+          SIGNPATH_REQUEST_ID: ${{ steps.submit-inner-signing.outputs.signing-request-id }}
+        run: |
+          Get-SignedArtifact `
+            -OrganizationId c37aa192-a27a-4377-9c90-5d6c95912dc0 `
+            -ApiToken $env:SIGNPATH_API_TOKEN `
+            -SigningRequestId $env:SIGNPATH_REQUEST_ID `
+            -OutputArtifactPath signed-inner.zip `
+            -Force `
+            -WaitForCompletionTimeoutInSeconds 3600
+
+          New-Item -ItemType Directory -Path signed-inner -Force
+          Expand-Archive -Path signed-inner.zip -DestinationPath signed-inner -Force
+
+      # Why: copy back strictly by the staged list so a layout mismatch in the
+      # returned artifact fails loudly (into fail-open) instead of silently
+      # shipping a mix of signed and unsigned binaries.
+      - name: Restore signed inner binaries into unpacked app
+        id: restore-signed-inner
+        if: matrix.platform == 'win' && steps.download-signed-inner.outcome == 'success'
+        continue-on-error: true
+        shell: pwsh
+        run: |
+          $root = Resolve-Path 'dist/win-unpacked'
+          $failures = New-Object System.Collections.Generic.List[string]
+          foreach ($relative in Get-Content 'inner-signing-list.txt') {
+            $signed = Get-ChildItem -Path signed-inner -Recurse -File |
+              Where-Object { [System.IO.Path]::GetRelativePath((Resolve-Path 'signed-inner'), $_.FullName).TrimStart('\', '/') -like "*$relative" } |
+              Select-Object -First 1
+            if ($null -eq $signed) {
+              $failures.Add("missing from signed artifact: $relative")
+              continue
+            }
+            $signature = Get-AuthenticodeSignature -FilePath $signed.FullName
+            if ($null -eq $signature.SignerCertificate) {
+              $failures.Add("returned without a signature: $relative")
+              continue
+            }
+            Copy-Item -Path $signed.FullName -Destination (Join-Path $root $relative) -Force
+            Write-Host ("{0,-14} {1}  <{2}>" -f $signature.Status, $relative, $signature.SignerCertificate.Subject)
+          }
+          if ($failures.Count -gt 0) {
+            $failures | ForEach-Object { Write-Host "::error::$_" }
+            throw "Signed inner artifact did not round-trip cleanly ($($failures.Count) failures)."
+          }
+
+      # Why this step exists: electron-builder's CopyElevateHelper re-copies a
+      # pristine elevate.exe from its download cache over resources\elevate.exe
+      # on EVERY nsis pack — including the --prepackaged rebuild below — which
+      # clobbered the SignPath signature in v1.4.129-rc.4. There is no supported
+      # way to disable just the copy, so we overwrite the cache's copy with our
+      # signed one (identical bytes plus signature) so the clobber becomes a
+      # no-op. Known quirk: the cache persists across releases via actions/cache,
+      # so later runs may see elevate.exe as already signed and skip staging it —
+      # that is fine (the signature is timestamped) and the evidence gate checks
+      # elevate.exe in the shipped installer unconditionally. If this ever causes
+      # trouble, delete this step; the only effect is elevate.exe shipping
+      # unsigned again, which the evidence gate will flag.
+      - name: Replace cached elevate.exe with the signed copy
+        id: sign-elevate-cache
+        if: matrix.platform == 'win' && steps.restore-signed-inner.outcome == 'success'
+        continue-on-error: true
+        shell: pwsh
+        run: |
+          $signed = 'dist/win-unpacked/resources/elevate.exe'
+          if (-not (Test-Path $signed)) {
+            Write-Host '::warning::No elevate.exe in win-unpacked resources; nothing to protect from the rebuild clobber.'
+            exit 0
+          }
+          $signature = Get-AuthenticodeSignature -FilePath $signed
+          $subject = if ($null -eq $signature.SignerCertificate) { '<none>' } else { $signature.SignerCertificate.Subject }
+          if ($signature.Status -ne 'Valid' -or $subject -notlike '*CN=SignPath Foundation*') {
+            Write-Host "::warning::win-unpacked elevate.exe is not SignPath-signed ($($signature.Status), $subject); skipping cache swap."
+            exit 0
+          }
+          $cached = @(Get-ChildItem "$env:LOCALAPPDATA\electron-builder\Cache\nsis" -Recurse -Filter elevate.exe -ErrorAction SilentlyContinue)
+          if ($cached.Count -eq 0) {
+            Write-Host '::warning::No cached elevate.exe found (electron-builder cache layout changed?); the rebuild will pack the unsigned copy and the evidence gate will flag it.'
+            exit 0
+          }
+          foreach ($file in $cached) {
+            Copy-Item -Path $signed -Destination $file.FullName -Force
+            Write-Host "Replaced $($file.FullName) with the SignPath-signed copy."
+          }
+
+      - name: Rebuild NSIS installer from signed unpacked app
+        id: rebuild-nsis-signed
+        if: matrix.platform == 'win' && steps.restore-signed-inner.outcome == 'success'
+        continue-on-error: true
+        shell: pwsh
+        run: |
+          # Why: keep the pre-rebuild artifacts so a failed rebuild can fall
+          # back to shipping them unchanged (fail-open).
+          New-Item -ItemType Directory -Path prepack-backup -Force | Out-Null
+          Copy-Item 'dist/orca-windows-setup.exe' 'prepack-backup/orca-windows-setup.exe' -Force
+          Copy-Item 'dist/latest.yml' 'prepack-backup/latest.yml' -Force
+
+          pnpm exec electron-builder --config config/electron-builder.config.cjs --win --publish never --prepackaged "$env:GITHUB_WORKSPACE\dist\win-unpacked"
+          if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }
+          if (-not (Test-Path 'dist/orca-windows-setup.exe')) {
+            throw 'electron-builder --prepackaged did not produce dist/orca-windows-setup.exe'
+          }
+
+      - name: Roll back to original installer after failed rebuild
+        if: matrix.platform == 'win' && steps.rebuild-nsis-signed.outcome == 'failure'
+        shell: pwsh
+        run: |
+          if (Test-Path 'prepack-backup/orca-windows-setup.exe') {
+            Copy-Item 'prepack-backup/orca-windows-setup.exe' 'dist/orca-windows-setup.exe' -Force
+            Copy-Item 'prepack-backup/latest.yml' 'dist/latest.yml' -Force
+            Write-Warning 'Restored pre-rebuild installer; this release ships with unsigned inner binaries.'
+          }
+      # ── End Windows inner-binary signing ───────────────────────────────
+
       - name: Upload unsigned Windows installer for SignPath
         if: matrix.platform == 'win'
         id: upload-unsigned-windows-installer
@@ -968,6 +1234,7 @@ jobs:
           SIGNPATH_REQUEST_URL: ${{ steps.submit-signing-request.outputs.signing-request-web-url }}
           TAG: ${{ needs.cut.outputs.tag }}
           GITHUB_RUN_URL: https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}
+          INNER_SIGNING_SUBMITTED: ${{ steps.submit-inner-signing.outcome == 'success' }}
         run: |
           if ([string]::IsNullOrWhiteSpace($env:SLACK_WEBHOOK_URL)) {
             throw 'SLACK_WEBHOOK_URL secret is required so release approvers know when SignPath is waiting.'
@@ -978,7 +1245,9 @@ jobs:
             $requestUrl = "https://app.signpath.io/Web/$env:SIGNPATH_ORGANIZATION_ID/SigningRequests/$env:SIGNPATH_REQUEST_ID"
           }
 
-          $message = "Orca Windows release $env:TAG is ready for SignPath approval.`n<$requestUrl|Open SignPath signing request>`n<$env:GITHUB_RUN_URL|Open GitHub Actions run>"
+          # Why: releases where inner signing fell through have only this one request.
+          $stage = if ($env:INNER_SIGNING_SUBMITTED -eq 'true') { 'installer signing request (2 of 2)' } else { 'signing request' }
+          $message = "Orca Windows release $env:TAG $stage is ready for SignPath approval.`n<$requestUrl|Open SignPath signing request>`n<$env:GITHUB_RUN_URL|Open GitHub Actions run>"
           $payload = @{
             text = $message
             blocks = @(
@@ -1061,6 +1330,90 @@ jobs:
           }
           $signature.SignerCertificate | Format-List Subject,Issuer,NotBefore,NotAfter,Thumbprint
 
+      # Why: evidence gate for inner-binary signing (issue #7785, supersedes
+      # PR #7170's Orca.exe-only gate — this covers every staged .exe/.dll/.node
+      # by extracting the shipped installer). Warn-only until the flow has been
+      # proven on a real release, then flip ORCA_WINDOWS_INNER_SIGNATURE_REQUIRED
+      # to 'true' so unsigned inner binaries block the release.
+      - name: Verify Windows inner binary signatures
+        if: matrix.platform == 'win'
+        shell: pwsh
+        env:
+          ORCA_WINDOWS_INNER_SIGNATURE_REQUIRED: 'false'
+          INNER_SIGNING_COMPLETED: ${{ steps.rebuild-nsis-signed.outcome == 'success' }}
+        run: |
+          $required = $env:ORCA_WINDOWS_INNER_SIGNATURE_REQUIRED -eq 'true'
+          if ($env:INNER_SIGNING_COMPLETED -ne 'true') {
+            $message = 'Windows inner-binary signing did not complete; this release ships unsigned inner binaries (fail-open, issue #7785).'
+            if ($required) { throw $message }
+            Write-Host "::warning::$message"
+            exit 0
+          }
+
+          # Why try/catch: while the gate is warn-only, even an unexpected
+          # script error (extraction hiccup, missing file) must not block
+          # the release — only the flip to required makes failures fatal.
+          try {
+            $report = New-Object System.Collections.Generic.List[string]
+            $failures = New-Object System.Collections.Generic.List[string]
+
+            # Why: verify the files a user actually gets on disk, not the build
+            # tree — 7z parses the NSIS exe directly as its embedded payload.
+            $7za = 'node_modules/7zip-bin/win/x64/7za.exe'
+            New-Item -ItemType Directory -Path inner-evidence-extract -Force | Out-Null
+            & $7za x 'dist/orca-windows-setup.exe' '-oinner-evidence-extract' -y | Out-Null
+
+            $root = Resolve-Path 'inner-evidence-extract'
+            # Why elevate.exe is always appended: staging skips already-signed
+            # files, and the persisted electron-builder cache can carry a
+            # previously signed elevate.exe — so it may be absent from the list
+            # in some runs, yet it is the file most at risk of losing its
+            # signature in the NSIS rebuild. Verify it in every release.
+            $targets = @(Get-Content 'inner-signing-list.txt')
+            if ($targets -notcontains 'resources\elevate.exe') {
+              $targets += 'resources\elevate.exe'
+            }
+            foreach ($relative in $targets) {
+              $path = Join-Path $root $relative
+              if (-not (Test-Path $path)) {
+                $failures.Add("missing from installer payload: $relative")
+                continue
+              }
+              $signature = Get-AuthenticodeSignature -FilePath $path
+              $subject = if ($null -eq $signature.SignerCertificate) { '<none>' } else { $signature.SignerCertificate.Subject }
+              $line = "{0,-14} {1}  <{2}>" -f $signature.Status, $relative, $subject
+              $report.Add($line)
+              Write-Host $line
+              if ($signature.Status -ne 'Valid' -or $subject -notlike '*CN=SignPath Foundation*') {
+                $failures.Add("not signed by SignPath Foundation: $relative ($($signature.Status), $subject)")
+              }
+            }
+
+            Set-Content -Path 'inner-signing-evidence.txt' -Value ($report -join "`n")
+            if ($failures.Count -gt 0) {
+              $failures | ForEach-Object { Write-Host "::warning::$_" }
+              $message = "Windows inner-binary evidence gate found $($failures.Count) problems."
+              if ($required) { throw $message }
+              Write-Host "::warning::$message Fail-open until ORCA_WINDOWS_INNER_SIGNATURE_REQUIRED is 'true'."
+            } else {
+              Write-Host "All $($targets.Count) inner binaries in the shipped installer are signed by SignPath Foundation."
+            }
+          } catch {
+            if ($required) { throw }
+            Write-Host "::warning::Windows inner-binary evidence gate errored: $_ (fail-open, issue #7785)."
+          }
+
+      - name: Upload Windows inner signing evidence
+        if: always() && matrix.platform == 'win'
+        uses: actions/upload-artifact@v7
+        with:
+          name: orca-windows-inner-signing-evidence-${{ needs.cut.outputs.tag }}
+          path: |
+            inner-signing-evidence.txt
+            inner-signing-list.txt
+          if-no-files-found: ignore
+          retention-days: 30
+
       - name: Publish signed Windows release artifacts
         if: matrix.platform == 'win'
         uses: nick-fields/retry@v4

.github/workflows/win-update-e2e.yml

@@ -0,0 +1,145 @@
+name: Windows Update-Survival E2E
+
+# Why: proves what happens to terminal sessions across a real Windows app update
+# by installing one released version, updating to another, and asserting on the
+# packaged artifacts (daemon survival, terminal interactivity, zero console
+# flashes). This MUST run on a disposable CI Windows: electron-builder's oneClick
+# installer uninstalls the registry-registered copy of the app before every
+# install, so running the harness on a machine with a real Orca install would
+# delete it. A fresh runner has no Orca registered, so it is the only safe home
+# for install-based testing. Manual dispatch only.
+
+# Why: workflow_dispatch only works once a workflow is on the default branch.
+# To exercise this from the unmerged feature branch WITHOUT touching main, a
+# push trigger scoped to that exact branch runs the workflow from the branch's
+# own tree. The paths filter keeps it to harness/workflow edits so ordinary
+# commits don't spend 30 min of Windows runner time. Push runs have no inputs,
+# so every parameter falls back to a default below.
+on:
+  push:
+    branches:
+      - Jinwoo-H/windows-update-survival
+    paths:
+      - 'tools/win-update-e2e/**'
+      - '.github/workflows/win-update-e2e.yml'
+  workflow_dispatch:
+    inputs:
+      from_tag:
+        description: Release tag to install first (e.g. v1.4.124-rc.8)
+        required: true
+        type: string
+      to_tag:
+        description: Release tag to update to (e.g. v1.4.124-rc.9)
+        required: true
+        type: string
+      expect:
+        description: Expected outcome profile
+        required: true
+        type: choice
+        default: cold-restore
+        options:
+          - cold-restore
+          - survival
+      asset_pattern:
+        description: Installer asset glob
+        required: false
+        type: string
+        default: '*windows-setup.exe'
+      soak_seconds:
+        description: Post-relaunch console-window watch duration
+        required: false
+        type: string
+        default: '60'
+
+permissions:
+  contents: read
+
+# Why: cancel a superseded run when iterating — a new push to the branch makes
+# the in-flight Windows job obsolete, so free the runner immediately.
+concurrency:
+  group: win-update-e2e-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  update-e2e:
+    name: update ${{ inputs.from_tag || 'v1.4.124-rc.8' }} -> ${{ inputs.to_tag || 'v1.4.124-rc.9' }} (${{ inputs.expect || 'cold-restore' }})
+    runs-on: windows-2022
+    timeout-minutes: 30
+    env:
+      FROM_TAG: ${{ inputs.from_tag || 'v1.4.124-rc.8' }}
+      TO_TAG: ${{ inputs.to_tag || 'v1.4.124-rc.9' }}
+      EXPECT: ${{ inputs.expect || 'cold-restore' }}
+      ASSET_PATTERN: ${{ inputs.asset_pattern || '*windows-setup.exe' }}
+      SOAK_SECONDS: ${{ inputs.soak_seconds || '60' }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+        with:
+          # This job only reads the repo and downloads a release; it never pushes.
+          persist-credentials: false
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version-file: package.json
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v6
+        with:
+          run_install: false
+
+      # Why: the harness only needs its runtime deps (the Playwright Electron
+      # driver); it drives already-built installer artifacts, so no app build.
+      - name: Install dependencies
+        run: pnpm install --frozen-lockfile
+
+      # Why: fail fast with a readable message if a tag has no matching Windows
+      # installer, rather than deep inside the harness.
+      - name: Download installers
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        shell: pwsh
+        run: |
+          New-Item -ItemType Directory -Force artifacts/from | Out-Null
+          New-Item -ItemType Directory -Force artifacts/to | Out-Null
+          gh release download "$env:FROM_TAG" --repo "${{ github.repository }}" --pattern "$env:ASSET_PATTERN" --dir artifacts/from
+          gh release download "$env:TO_TAG" --repo "${{ github.repository }}" --pattern "$env:ASSET_PATTERN" --dir artifacts/to
+          $from = Get-ChildItem artifacts/from -Filter *.exe | Select-Object -First 1
+          $to = Get-ChildItem artifacts/to -Filter *.exe | Select-Object -First 1
+          if (-not $from) { throw "No installer matching '$env:ASSET_PATTERN' in $env:FROM_TAG" }
+          if (-not $to) { throw "No installer matching '$env:ASSET_PATTERN' in $env:TO_TAG" }
+          "FROM_EXE=$($from.FullName)" | Out-File -FilePath $env:GITHUB_ENV -Encoding utf8 -Append
+          "TO_EXE=$($to.FullName)" | Out-File -FilePath $env:GITHUB_ENV -Encoding utf8 -Append
+
+      # Why: no --install-dir here. A CI runner has no real Orca registered, so
+      # the harness installs to the default per-user location and owns it — the
+      # isolated-install machinery exists only to (imperfectly) protect a real
+      # developer install, which does not apply on a throwaway runner.
+      - name: Run update-survival harness
+        id: harness
+        shell: pwsh
+        env:
+          # Why: on a driving failure the harness dumps a screenshot + DOM state
+          # here so the actual post-onboarding UI is visible in the artifacts.
+          ORCA_E2E_DIAG_DIR: artifacts/diag
+        run: |
+          $log = "artifacts/harness-output.log"
+          node tools/win-update-e2e/run.mjs `
+            --from "$env:FROM_EXE" `
+            --to "$env:TO_EXE" `
+            --expect "$env:EXPECT" `
+            --soak-seconds "$env:SOAK_SECONDS" 2>&1 | Tee-Object -FilePath $log
+          exit $LASTEXITCODE
+
+      - name: Upload harness output
+        if: always()
+        uses: actions/upload-artifact@v7
+        with:
+          name: win-update-e2e-output
+          # Why: log + diagnostics only, never the multi-hundred-MB installers.
+          path: |
+            artifacts/harness-output.log
+            artifacts/diag/**
+          retention-days: 7
+          if-no-files-found: warn

.github/workflows/win-update-survival-e2e.yml

@@ -0,0 +1,116 @@
+name: Windows Update-Survival E2E (branch build)
+
+# Why: proves the Phase 1 daemon-host relocation actually makes terminal
+# sessions SURVIVE a Windows update. Builds an (unsigned) installer FROM THIS
+# BRANCH, then runs the win-update-e2e harness with --expect survival: install,
+# open a terminal, silently update over it, relaunch, and assert the daemon PID
+# is unchanged and the pre-update terminal is still interactive with no console
+# flashing. Runs from the feature branch via push (workflow_dispatch only works
+# on the default branch); main is never touched. A CI runner is the only safe
+# place to install/update — see the relocation post-mortem.
+
+on:
+  push:
+    branches:
+      - Jinwoo-H/windows-update-survival
+    paths:
+      - 'src/main/daemon/**'
+      - 'src/main/pty/**'
+      - 'tools/win-update-e2e/**'
+      - 'config/electron-builder.config.cjs'
+      - '.github/workflows/win-update-survival-e2e.yml'
+  workflow_dispatch:
+    inputs:
+      expect:
+        description: Expected outcome profile
+        required: true
+        type: choice
+        default: survival
+        options:
+          - survival
+          - cold-restore
+
+permissions:
+  contents: read
+
+concurrency:
+  group: win-update-survival-e2e-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  survival:
+    name: survival (branch build over itself)
+    runs-on: windows-2022
+    timeout-minutes: 50
+    env:
+      EXPECT: ${{ inputs.expect || 'survival' }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+        with:
+          # This job only builds and runs the harness locally; it never pushes.
+          persist-credentials: false
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version-file: package.json
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v6
+        with:
+          run_install: false
+
+      - name: Install dependencies
+        run: pnpm install --frozen-lockfile
+
+      # Why: cache the built installer keyed on the inputs that affect it, so a
+      # harness-only edit skips the ~20 min electron-builder build. The daemon
+      # relocation code lives under src/, so changing it correctly rebuilds.
+      - name: Cache branch installer
+        id: cache-installer
+        uses: actions/cache@v4
+        with:
+          path: dist/orca-windows-setup.exe
+          key: branch-installer-${{ hashFiles('src/**', 'config/**', 'package.json', 'pnpm-lock.yaml') }}
+
+      - name: Build Windows installer (unsigned)
+        if: steps.cache-installer.outputs.cache-hit != 'true'
+        run: |
+          node config/scripts/ensure-native-runtime.mjs --runtime=electron
+          pnpm run build:desktop
+          pnpm exec electron-builder --config config/electron-builder.config.cjs --win --publish never
+
+      # Why: install the branch build, open a terminal, update the SAME build
+      # over it, and assert the relocated daemon survives. Same build on both
+      # sides isolates the survival mechanism (NSIS kill sweep + userData daemon
+      # + adoption) from cross-version staleness policy. No --install-dir: a CI
+      # runner has no real Orca to protect.
+      - name: Run survival harness
+        id: harness
+        shell: pwsh
+        env:
+          ORCA_E2E_DIAG_DIR: artifacts/diag
+        run: |
+          New-Item -ItemType Directory -Force artifacts | Out-Null
+          $exe = "dist/orca-windows-setup.exe"
+          if (-not (Test-Path $exe)) { throw "Installer not found at $exe" }
+          $log = "artifacts/survival-output.log"
+          node tools/win-update-e2e/run.mjs `
+            --from "$exe" `
+            --to "$exe" `
+            --expect "$env:EXPECT" `
+            --soak-seconds 60 2>&1 | Tee-Object -FilePath $log
+          exit $LASTEXITCODE
+
+      - name: Upload survival output
+        if: always()
+        uses: actions/upload-artifact@v7
+        with:
+          name: win-update-survival-output
+          path: |
+            artifacts/survival-output.log
+            artifacts/diag/**
+          retention-days: 7
+          if-no-files-found: warn

.github/workflows/windows-signing-rehearsal.yml

@@ -0,0 +1,360 @@
+# Windows inner-binary signing rehearsal.
+#
+# Why: SignPath cannot deep-sign inside NSIS installers, so shipping signed
+# inner binaries (Orca.exe, node-pty *.node, DLLs — see issue #7785) requires
+# a two-request flow: sign the unpacked PE files first, then build the NSIS
+# installer from the signed tree, then sign the installer. This workflow
+# rehearses that entire flow from a branch, end to end, without publishing
+# anything — so the release pipeline on main is never at risk while we verify.
+#
+# Runs only via manual dispatch. Use the test-signing policy for iteration
+# (auto-approved test certificate) and release-signing to rehearse the
+# production flow (requires manual SignPath approvals). Rehearsed successfully
+# with test-signing (run 28987534795) and release-signing (run 28988432001).
+name: Windows signing rehearsal
+
+on:
+  workflow_dispatch:
+    inputs:
+      signing-policy-slug:
+        description: SignPath signing policy
+        type: choice
+        default: test-signing
+        options:
+          - test-signing
+          - release-signing
+      inner-artifact-configuration-slug:
+        description: SignPath artifact configuration for the inner-binaries zip
+        type: string
+        default: windows-inner-binaries-zip
+
+jobs:
+  rehearse:
+    # Why: SignPath origin verification requires GitHub-hosted runners, and
+    # windows-2022 matches the real release job's build environment.
+    runs-on: windows-2022
+    timeout-minutes: 360
+    permissions:
+      actions: read
+      contents: read
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+        with:
+          # Why: nothing here pushes; keep the token out of .git/config.
+          persist-credentials: false
+
+      - name: Setup pnpm
+        uses: pnpm/action-setup@v6
+        with:
+          run_install: false
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v6
+        with:
+          node-version-file: package.json
+          cache: pnpm
+
+      - name: Cache electron-builder downloads
+        uses: actions/cache@v5
+        with:
+          path: |
+            ~\AppData\Local\electron\Cache
+            ~\AppData\Local\electron-builder\Cache
+          key: electron-builder-win-${{ hashFiles('pnpm-lock.yaml') }}
+          restore-keys: |
+            electron-builder-win-
+
+      - name: Install dependencies
+        uses: nick-fields/retry@v4
+        with:
+          timeout_minutes: 10
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: pnpm install --frozen-lockfile
+
+      # Why: rehearsal builds are never published, so the official-build
+      # secrets (telemetry key, diagnostics URL) are intentionally omitted.
+      - name: Build app
+        run: pnpm build:release
+        env:
+          NODE_OPTIONS: --max-old-space-size=4096
+
+      - name: Package unpacked Windows app
+        shell: pwsh
+        run: |
+          node config/scripts/ensure-native-runtime.mjs --runtime=electron
+          if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }
+          pnpm exec electron-builder --config config/electron-builder.config.cjs --win --dir --publish never
+          if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }
+          if (-not (Test-Path 'dist/win-unpacked/Orca.exe')) {
+            throw 'electron-builder --dir did not produce dist/win-unpacked/Orca.exe'
+          }
+
+      # Why: only unsigned PE files go to SignPath. Files that already carry a
+      # valid signature (Microsoft's OpenConsole.exe, signed vendor DLLs) must
+      # keep their original signer, so they are excluded from the request.
+      - name: Stage unsigned PE files for inner signing
+        id: stage-inner
+        shell: pwsh
+        run: |
+          $root = Resolve-Path 'dist/win-unpacked'
+          $stage = New-Item -ItemType Directory -Force -Path 'signing-stage'
+          $list = New-Object System.Collections.Generic.List[string]
+          $skipped = New-Object System.Collections.Generic.List[string]
+
+          Get-ChildItem -Path $root -Recurse -File |
+            Where-Object { $_.Extension -in '.exe', '.dll', '.node' } |
+            ForEach-Object {
+              $relative = [System.IO.Path]::GetRelativePath($root, $_.FullName)
+              $signature = Get-AuthenticodeSignature -FilePath $_.FullName
+              if ($signature.Status -eq 'Valid') {
+                $skipped.Add("$relative  <already signed: $($signature.SignerCertificate.Subject)>")
+                return
+              }
+              $destination = Join-Path $stage.FullName $relative
+              New-Item -ItemType Directory -Force -Path (Split-Path $destination) | Out-Null
+              Copy-Item -Path $_.FullName -Destination $destination -Force
+              $list.Add($relative)
+            }
+
+          if (-not ($list -contains 'Orca.exe')) {
+            throw 'Orca.exe was not staged for signing; unpacked layout changed?'
+          }
+          if (-not ($list | Where-Object { $_ -like '*conpty_console_list.node' })) {
+            throw 'node-pty conpty_console_list.node was not staged; this is the file from issue #7785.'
+          }
+
+          Set-Content -Path 'inner-signing-list.txt' -Value ($list -join "`n")
+          Write-Host "Staged $($list.Count) unsigned PE files for signing:"
+          $list | ForEach-Object { Write-Host "  $_" }
+          Write-Host "Skipped $($skipped.Count) already-signed files:"
+          $skipped | ForEach-Object { Write-Host "  $_" }
+
+      - name: Upload unsigned inner binaries for SignPath
+        id: upload-unsigned-inner
+        uses: actions/upload-artifact@v7
+        with:
+          name: orca-windows-inner-unsigned-${{ github.run_id }}
+          path: signing-stage/**
+          if-no-files-found: error
+
+      - name: Install SignPath PowerShell module
+        shell: pwsh
+        run: |
+          $ErrorActionPreference = 'Stop'
+          [Net.ServicePointManager]::SecurityProtocol = [Net.ServicePointManager]::SecurityProtocol -bor [Net.SecurityProtocolType]::Tls12
+          $useResourceGet = $null -ne (Get-Command -Name Install-PSResource -ErrorAction SilentlyContinue)
+          if ($useResourceGet) {
+            if ($null -eq (Get-PSResourceRepository -Name PSGallery -ErrorAction SilentlyContinue)) {
+              Register-PSResourceRepository -PSGallery -Trusted
+            } else {
+              Set-PSResourceRepository -Name PSGallery -Trusted
+            }
+            Install-PSResource -Name SignPath -Version '[4.0.0,5.0.0)' -Repository PSGallery -Scope CurrentUser -TrustRepository -Reinstall -ErrorAction Stop
+          } else {
+            Install-Module -Name SignPath -Repository PSGallery -MinimumVersion 4.0.0 -MaximumVersion 4.999.999 -Scope CurrentUser -Force -AllowClobber -ErrorAction Stop
+          }
+          Import-Module SignPath -ErrorAction Stop
+          Get-Command -Name Get-SignedArtifact -Module SignPath -ErrorAction Stop
+
+      - name: Submit inner binaries signing request
+        id: submit-inner-signing
+        uses: signpath/github-action-submit-signing-request@v2
+        with:
+          api-token: ${{ secrets.SIGNPATH_API_TOKEN }}
+          organization-id: c37aa192-a27a-4377-9c90-5d6c95912dc0
+          project-slug: orca
+          signing-policy-slug: ${{ inputs.signing-policy-slug || 'test-signing' }}
+          artifact-configuration-slug: ${{ inputs.inner-artifact-configuration-slug || 'windows-inner-binaries-zip' }}
+          github-artifact-id: ${{ steps.upload-unsigned-inner.outputs.artifact-id }}
+          wait-for-completion: false
+
+      - name: Download signed inner binaries from SignPath
+        shell: pwsh
+        env:
+          SIGNPATH_API_TOKEN: ${{ secrets.SIGNPATH_API_TOKEN }}
+          SIGNPATH_REQUEST_ID: ${{ steps.submit-inner-signing.outputs.signing-request-id }}
+        run: |
+          Get-SignedArtifact `
+            -ApiToken $env:SIGNPATH_API_TOKEN `
+            -OrganizationId 'c37aa192-a27a-4377-9c90-5d6c95912dc0' `
+            -SigningRequestId $env:SIGNPATH_REQUEST_ID `
+            -OutputArtifactPath signed-inner.zip `
+            -Force `
+            -WaitForCompletionTimeoutInSeconds 3600
+          New-Item -ItemType Directory -Path signed-inner -Force
+          Expand-Archive -Path signed-inner.zip -DestinationPath signed-inner -Force
+
+      # Why: copy back strictly by the staged list so a layout mismatch in the
+      # returned artifact fails loudly instead of silently shipping a mix of
+      # signed and unsigned binaries.
+      - name: Restore signed inner binaries into unpacked app
+        shell: pwsh
+        run: |
+          $root = Resolve-Path 'dist/win-unpacked'
+          $failures = New-Object System.Collections.Generic.List[string]
+          foreach ($relative in Get-Content 'inner-signing-list.txt') {
+            $signed = Get-ChildItem -Path signed-inner -Recurse -File |
+              Where-Object { [System.IO.Path]::GetRelativePath((Resolve-Path 'signed-inner'), $_.FullName).TrimStart('\', '/') -like "*$relative" } |
+              Select-Object -First 1
+            if ($null -eq $signed) {
+              $failures.Add("missing from signed artifact: $relative")
+              continue
+            }
+            $signature = Get-AuthenticodeSignature -FilePath $signed.FullName
+            if ($null -eq $signature.SignerCertificate) {
+              $failures.Add("returned without a signature: $relative")
+              continue
+            }
+            Copy-Item -Path $signed.FullName -Destination (Join-Path $root $relative) -Force
+            Write-Host ("{0,-14} {1}  <{2}>" -f $signature.Status, $relative, $signature.SignerCertificate.Subject)
+          }
+          if ($failures.Count -gt 0) {
+            $failures | ForEach-Object { Write-Host "::error::$_" }
+            throw "Signed inner artifact did not round-trip cleanly ($($failures.Count) failures)."
+          }
+
+      - name: Build NSIS installer from signed unpacked app
+        shell: pwsh
+        run: |
+          pnpm exec electron-builder --config config/electron-builder.config.cjs --win --publish never --prepackaged "$env:GITHUB_WORKSPACE\dist\win-unpacked"
+          if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }
+          if (-not (Test-Path 'dist/orca-windows-setup.exe')) {
+            throw 'electron-builder --prepackaged did not produce dist/orca-windows-setup.exe'
+          }
+
+      - name: Upload unsigned Windows installer for SignPath
+        id: upload-unsigned-installer
+        uses: actions/upload-artifact@v7
+        with:
+          name: orca-windows-installer-unsigned-${{ github.run_id }}
+          path: dist/orca-windows-setup.exe
+          if-no-files-found: error
+
+      - name: Submit Windows installer signing request
+        id: submit-installer-signing
+        uses: signpath/github-action-submit-signing-request@v2
+        with:
+          api-token: ${{ secrets.SIGNPATH_API_TOKEN }}
+          organization-id: c37aa192-a27a-4377-9c90-5d6c95912dc0
+          project-slug: orca
+          signing-policy-slug: ${{ inputs.signing-policy-slug || 'test-signing' }}
+          artifact-configuration-slug: github-actions-windows-installer
+          github-artifact-id: ${{ steps.upload-unsigned-installer.outputs.artifact-id }}
+          wait-for-completion: false
+
+      - name: Download signed Windows installer from SignPath
+        shell: pwsh
+        env:
+          SIGNPATH_API_TOKEN: ${{ secrets.SIGNPATH_API_TOKEN }}
+          SIGNPATH_REQUEST_ID: ${{ steps.submit-installer-signing.outputs.signing-request-id }}
+        run: |
+          Get-SignedArtifact `
+            -ApiToken $env:SIGNPATH_API_TOKEN `
+            -OrganizationId 'c37aa192-a27a-4377-9c90-5d6c95912dc0' `
+            -SigningRequestId $env:SIGNPATH_REQUEST_ID `
+            -OutputArtifactPath signed-windows.zip `
+            -Force `
+            -WaitForCompletionTimeoutInSeconds 14400
+          New-Item -ItemType Directory -Path signed-windows -Force
+          Expand-Archive -Path signed-windows.zip -DestinationPath signed-windows -Force
+
+      # Why: signing changes installer bytes, so the updater metadata must be
+      # regenerated exactly like the release job's staging step does.
+      - name: Stage signed Windows installer and regenerate updater metadata
+        shell: pwsh
+        run: |
+          $signedInstaller = Get-ChildItem -Path signed-windows -Recurse -File -Filter 'orca-windows-setup.exe' | Select-Object -First 1
+          if ($null -eq $signedInstaller) {
+            throw 'Signed Windows installer was not returned by SignPath.'
+          }
+          Copy-Item -Path $signedInstaller.FullName -Destination 'dist/orca-windows-setup.exe' -Force
+          & 'node_modules/app-builder-bin/win/x64/app-builder.exe' blockmap --input 'dist/orca-windows-setup.exe' --output 'dist/orca-windows-setup.exe.blockmap'
+          if ($LASTEXITCODE -ne 0) { exit $LASTEXITCODE }
+
+          $installer = Get-Item 'dist/orca-windows-setup.exe'
+          $blockmap = Get-Item 'dist/orca-windows-setup.exe.blockmap'
+          $stream = [System.IO.File]::OpenRead($installer.FullName)
+          try {
+            $sha512 = [System.Security.Cryptography.SHA512]::Create()
+            $hash = [Convert]::ToBase64String($sha512.ComputeHash($stream))
+          } finally {
+            if ($null -ne $sha512) { $sha512.Dispose() }
+            $stream.Dispose()
+          }
+          $latestYml = Get-Content -Path 'dist/latest.yml' -Raw
+          $latestYml = [regex]::Replace($latestYml, '(?m)^(\s*)sha512: .+$', {
+            param($match)
+            "$($match.Groups[1].Value)sha512: $hash"
+          })
+          $latestYml = $latestYml -replace '(?m)^    size: \d+$', "    size: $($installer.Length)"
+          $latestYml = $latestYml -replace '(?m)^    blockMapSize: \d+$', "    blockMapSize: $($blockmap.Length)"
+          Set-Content -Path 'dist/latest.yml' -Value $latestYml -NoNewline
+
+      # Why: this is the pass/fail heart of the rehearsal — the same evidence
+      # check that will later gate releases (see PR #7170). Test certificates
+      # do not chain to a trusted root, so Status=Valid is only required for
+      # release-signing runs; test-signing runs require a present signature.
+      - name: Verify signatures end to end
+        shell: pwsh
+        env:
+          SIGNING_POLICY: ${{ inputs.signing-policy-slug || 'test-signing' }}
+        run: |
+          $report = New-Object System.Collections.Generic.List[string]
+          $failures = New-Object System.Collections.Generic.List[string]
+          $requireValid = $env:SIGNING_POLICY -eq 'release-signing'
+
+          function Test-Signature([string]$label, [string]$path) {
+            $signature = Get-AuthenticodeSignature -FilePath $path
+            $subject = if ($null -eq $signature.SignerCertificate) { '<none>' } else { $signature.SignerCertificate.Subject }
+            $line = "{0,-14} {1}  <{2}>" -f $signature.Status, $label, $subject
+            $script:report.Add($line)
+            Write-Host $line
+            if ($null -eq $signature.SignerCertificate -or $signature.Status -eq 'NotSigned') {
+              $script:failures.Add("unsigned: $label")
+            } elseif ($script:requireValid -and $signature.Status -ne 'Valid') {
+              $script:failures.Add("not Valid under release-signing: $label ($($signature.Status))")
+            } elseif ($script:requireValid -and $subject -notlike '*CN=SignPath Foundation*') {
+              $script:failures.Add("unexpected signer: $label ($subject)")
+            }
+          }
+
+          Test-Signature 'installer: orca-windows-setup.exe' 'dist/orca-windows-setup.exe'
+
+          # Extract the signed installer and verify the files a user actually
+          # gets on disk — including the exact file from issue #7785.
+          $7za = 'node_modules/7zip-bin/win/x64/7za.exe'
+          New-Item -ItemType Directory -Path extracted-app -Force | Out-Null
+          & $7za x 'dist/orca-windows-setup.exe' '-oextracted-app' -y | Out-Null
+
+          $root = Resolve-Path 'extracted-app'
+          foreach ($relative in Get-Content 'inner-signing-list.txt') {
+            $path = Join-Path $root $relative
+            if (-not (Test-Path $path)) {
+              $failures.Add("missing from installer payload: $relative")
+              continue
+            }
+            Test-Signature "installed: $relative" $path
+          }
+
+          Set-Content -Path 'signing-evidence.txt' -Value ($report -join "`n")
+          if ($failures.Count -gt 0) {
+            $failures | ForEach-Object { Write-Host "::error::$_" }
+            throw "Signing rehearsal failed with $($failures.Count) problems."
+          }
+          Write-Host "All $((Get-Content 'inner-signing-list.txt').Count) inner binaries plus the installer are signed."
+
+      - name: Upload rehearsal evidence and installer
+        if: always()
+        uses: actions/upload-artifact@v7
+        with:
+          name: windows-signing-rehearsal-${{ inputs.signing-policy-slug || 'test-signing' }}-${{ github.run_id }}
+          path: |
+            signing-evidence.txt
+            inner-signing-list.txt
+            dist/orca-windows-setup.exe
+            dist/orca-windows-setup.exe.blockmap
+            dist/latest.yml
+          retention-days: 7

.gitignore

@@ -60,6 +60,10 @@ pnpm-debug.log*
 # Coverage
 coverage/
 
+# Benchmark run output
+/tools/benchmarks/results/*.json
+/.bench-fixtures/
+
 # Temp
 tmp/
 .tmp/

.oxlintrc.json

@@ -57,7 +57,9 @@
     "unicorn/no-array-reverse": "error",
     "unicorn/no-instanceof-builtins": "error",
     "unicorn/no-useless-promise-resolve-reject": "error",
+    "unicorn/prefer-array-find": "error",
     "unicorn/prefer-array-flat-map": "warn",
+    "unicorn/prefer-array-index-of": "error",
     "unicorn/prefer-at": "error",
     "unicorn/prefer-date-now": "error",
     "unicorn/prefer-includes": "error",
@@ -65,6 +67,7 @@
     "unicorn/prefer-math-min-max": "error",
     "unicorn/prefer-negative-index": "error",
     "unicorn/prefer-node-protocol": "error",
+    "unicorn/prefer-number-properties": "error",
     "unicorn/prefer-object-from-entries": "error",
     "unicorn/prefer-regexp-test": "warn",
     "unicorn/prefer-ternary": "error",

AGENTS.md

@@ -12,7 +12,9 @@ Keep comments short — one or two lines. Capture only the non-obvious reason (s
 
 ## Lint Rules: Do Not Disable Max Lines
 
-Never add a `max-lines` disable (`eslint-disable max-lines`, `oxlint-disable max-lines`, or line-specific variants). Split the file, extract focused modules, move fixtures/builders into named files, or otherwise reduce the counted lines instead.
+Never add a `max-lines` disable (`eslint-disable max-lines`, `oxlint-disable max-lines`, or line-specific variants), and never add a per-file `max-lines` bump in `mobile/.oxlintrc.json`. Split the file, extract focused modules, move fixtures/builders into named files, or otherwise reduce the counted lines instead.
+
+This is enforced in CI by `pnpm check:max-lines-ratchet` (`config/scripts/check-max-lines-ratchet.mjs`). Files already over the limit are grandfathered in `config/max-lines-baseline.txt`; that list may only **shrink**. Adding a new suppression fails the build. If you remove a suppression (great — split it!), run `pnpm check:max-lines-ratchet --prune` to drop its baseline line.
 
 ## File and Module Naming
 

README.md

@@ -12,7 +12,7 @@
 </p>
 
 <p align="center">
-  <sub><a href="docs/readme/README.es.md">Español</a> · <a href="docs/readme/README.zh-CN.md">中文</a> · <a href="docs/readme/README.ja.md">日本語</a> · <a href="docs/readme/README.ko.md">한국어</a></sub>
+  <sub><a href="docs/readme/README.es.md">Español</a> · <a href="docs/readme/README.pt.md">Português</a> · <a href="docs/readme/README.zh-CN.md">中文</a> · <a href="docs/readme/README.ja.md">日本語</a> · <a href="docs/readme/README.ko.md">한국어</a></sub>
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@
 
 Monitor and steer your agents from your phone — get notified when an agent finishes and send follow-ups from anywhere.
 
-[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [TestFlight](https://testflight.apple.com/join/YjeGMQBA) · [Android APK 0.0.21](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk) · [Docs →](https://www.onorca.dev/docs/mobile)
+[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [TestFlight](https://testflight.apple.com/join/YjeGMQBA) · [Android APK 0.0.25](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [Docs →](https://www.onorca.dev/docs/mobile)
 
 </td>
 <td width="50%">
@@ -230,7 +230,7 @@ yay -S stably-orca-bin
 Pair with your desktop app to monitor and steer your agents from your phone.
 
 - **iOS:** [Download on the App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) or [join TestFlight](https://testflight.apple.com/join/YjeGMQBA)
-- **Android:** [Download APK 0.0.21](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk)
+- **Android:** [Download APK 0.0.25](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
 
 ---
 
@@ -238,9 +238,10 @@ Pair with your desktop app to monitor and steer your agents from your phone.
 
 - **Discord:** Join the community on **[Discord](https://discord.gg/fzjDKHxv8Q)**.
 - **Twitter / X:** Follow **[@orca_build](https://x.com/orca_build)** for updates and announcements.
-- **WeChat:** Scan the QR code to join the community.
+- **WeChat:** Scan the QR code to join the community. If the first group is full, use the backup group.
 
   <img src="docs/assets/wechat-qr.png" alt="WeChat QR code for the Orca community" width="160" />
+  <img src="docs/assets/wechat-qr-backup.jpg" alt="Backup WeChat QR code for the Orca community" width="160" />
 
 - **Feedback &amp; Ideas:** We ship fast. Missing something? [Request a new feature](https://github.com/stablyai/orca/issues).
 - **Privacy:** See the [privacy &amp; telemetry docs](https://www.onorca.dev/docs/telemetry) for what anonymous usage data Orca collects and how to opt out.
@@ -256,6 +257,10 @@ Want to contribute or run locally? See our [CONTRIBUTING.md](.github/CONTRIBUTIN
   <img src="https://contrib.rocks/image?repo=stablyai/orca" alt="Orca contributors" />
 </a>
 
+<p align="center">
+  <img src="docs/assets/star-history.png" alt="GitHub star history chart for stablyai/orca" width="880" />
+</p>
+
 ## License
 
 Orca is free and open source under the [MIT License](LICENSE).

build-plugins/plain-node-entry-guard.ts

@@ -0,0 +1,130 @@
+import { spawnSync } from 'node:child_process'
+import { join } from 'node:path'
+import type { NormalizedOutputOptions, OutputBundle, OutputChunk, Plugin } from 'rollup'
+
+// Why: v1.4.129-rc.1 shipped a dead terminal daemon because a shared main
+// chunk gained `require("electron")` (an import edge added in #7642), and the
+// daemon is forked as a plain-Node process where electron cannot be required.
+// Nothing in CI executes the built daemon-entry under plain Node, so the leak
+// stayed invisible until an adopted old daemon died. This guard fails the
+// build when any chunk reachable from a plain-Node fork entry requires
+// electron, and smoke-loads daemon-entry under plain Node to prove its module
+// graph still resolves.
+
+// Entries executed as plain Node (ELECTRON_RUN_AS_NODE / no electron runtime):
+// forked daemon, parcel-watcher and computer sidecars, and the CLI-run
+// agent-hooks entry. require("electron") throws MODULE_NOT_FOUND in all of them.
+const PLAIN_NODE_ENTRY_NAMES = [
+  'daemon-entry',
+  'parcel-watcher-process-entry',
+  'computer-sidecar',
+  'agent-hooks/managed-agent-hook-controls'
+] as const
+
+const ELECTRON_REQUIRE_RE = /require\(\s*["']electron["']\s*\)/
+
+function collectReachableChunks(
+  entry: OutputChunk,
+  byFileName: Map<string, OutputChunk>
+): OutputChunk[] {
+  const seen = new Set<string>()
+  const reachable: OutputChunk[] = []
+  const stack = [entry.fileName]
+  while (stack.length > 0) {
+    const fileName = stack.pop() as string
+    if (seen.has(fileName)) {
+      continue
+    }
+    seen.add(fileName)
+    const chunk = byFileName.get(fileName)
+    if (!chunk) {
+      continue
+    }
+    reachable.push(chunk)
+    for (const imported of [...chunk.imports, ...chunk.dynamicImports]) {
+      stack.push(imported)
+    }
+  }
+  return reachable
+}
+
+function assertNoElectronRequire(
+  entryName: string,
+  entry: OutputChunk,
+  byFileName: Map<string, OutputChunk>
+): void {
+  for (const chunk of collectReachableChunks(entry, byFileName)) {
+    if (ELECTRON_REQUIRE_RE.test(chunk.code)) {
+      throw new Error(
+        `[plain-node-entry-guard] "${entryName}" reaches chunk "${chunk.fileName}" that ` +
+          `requires electron. "${entryName}" runs as a plain-Node process, where ` +
+          `require("electron") throws MODULE_NOT_FOUND and kills it at startup (the ` +
+          `v1.4.129-rc.1 daemon outage). Keep electron imports out of its module graph.`
+      )
+    }
+  }
+}
+
+// Why: proves the whole daemon-entry graph resolves under plain Node (no
+// unresolved requires). require("electron") does not throw in a dev tree with
+// node_modules present, so the static scan above — not this smoke — is the
+// electron regression guard; this only catches gross load failures.
+function smokeLoadDaemonEntry(outputDir: string): void {
+  const entryPath = join(outputDir, 'daemon-entry.js')
+  const result = spawnSync(process.execPath, [entryPath], {
+    encoding: 'utf8',
+    timeout: 15_000
+  })
+  if (result.error) {
+    throw new Error(
+      `[plain-node-entry-guard] could not smoke-load daemon-entry.js under plain Node: ` +
+        `${result.error.message}`
+    )
+  }
+  const stderr = result.stderr ?? ''
+  if (/Cannot find module|MODULE_NOT_FOUND/.test(stderr)) {
+    throw new Error(
+      `[plain-node-entry-guard] daemon-entry.js failed to load under plain Node:\n${stderr}`
+    )
+  }
+  if (!stderr.includes('Usage: daemon-entry')) {
+    throw new Error(
+      `[plain-node-entry-guard] daemon-entry.js did not reach argv parsing under plain Node ` +
+        `(expected the "Usage: daemon-entry" error). stderr:\n${stderr}`
+    )
+  }
+}
+
+export function createPlainNodeEntryGuardPlugin(): Plugin {
+  return {
+    name: 'orca-plain-node-entry-guard',
+    writeBundle(options: NormalizedOutputOptions, bundle: OutputBundle) {
+      // Why: skip in `electron-vite dev` watch mode — the smoke would respawn on
+      // every rebuild, and the guard only needs to gate produced builds.
+      if (this.meta.watchMode) {
+        return
+      }
+      const chunks = Object.values(bundle).filter(
+        (item): item is OutputChunk => item.type === 'chunk'
+      )
+      const byFileName = new Map(chunks.map((chunk) => [chunk.fileName, chunk]))
+      const entryByName = new Map<string, OutputChunk>()
+      for (const chunk of chunks) {
+        if (chunk.isEntry && chunk.name) {
+          entryByName.set(chunk.name, chunk)
+        }
+      }
+
+      for (const entryName of PLAIN_NODE_ENTRY_NAMES) {
+        const entry = entryByName.get(entryName)
+        if (entry) {
+          assertNoElectronRequire(entryName, entry, byFileName)
+        }
+      }
+
+      if (entryByName.has('daemon-entry') && options.dir) {
+        smokeLoadDaemonEntry(options.dir)
+      }
+    }
+  }
+}

config/electron-builder.config.cjs

@@ -2,6 +2,7 @@ const { chmodSync, existsSync, readdirSync } = require('node:fs')
 const { execFileSync } = require('node:child_process')
 const { join, resolve } = require('node:path')
 const electronBuilderNativeRebuild = require('./scripts/electron-builder-native-rebuild.cjs')
+const { verifyPackagedDaemonEntryBoots } = require('./scripts/verify-packaged-daemon-entry.cjs')
 const {
   createPackagedRuntimeNodeModuleResources,
   prunePackagedRuntimeNodeModules,
@@ -106,6 +107,7 @@ module.exports = {
     'out/main/win32-utils.js',
     'out/main/daemon-entry.js',
     'out/main/computer-sidecar.js',
+    'out/main/parcel-watcher-process-entry.js',
     'out/main/chunks/**',
     'resources/**',
     'node_modules/ws/**',
@@ -129,6 +131,20 @@ module.exports = {
     }
     prunePackagedRuntimeNodeModules(resourcesDir, context.electronPlatformName, context.arch)
     verifyPackagedMainRuntimeDeps(resourcesDir)
+    // Why: boot the packaged daemon-entry under plain Node, but only for the
+    // slice matching the packaging host's arch — daemon-entry.js is JS, yet it
+    // require()s the native (N-API) node-pty for the TARGET arch, which the host
+    // Node cannot load cross-arch. `Arch` enum: ia32=0, x64=1, armv7l=2,
+    // arm64=3, universal=4 (universal contains the host slice, so run it).
+    const archEnumByNodeArch = { ia32: 0, x64: 1, armv7l: 2, arm64: 3 }
+    const hostArchEnum = archEnumByNodeArch[process.arch]
+    if (context.arch === hostArchEnum || context.arch === 4) {
+      verifyPackagedDaemonEntryBoots(resourcesDir)
+    } else {
+      console.log(
+        `[verify-packaged-daemon-entry] skipped cross-arch slice (target ${context.arch}, host ${process.arch})`
+      )
+    }
     chmodUnixCliLaunchers(resourcesDir, context.electronPlatformName)
     chmodMacServeSimHelpers(resourcesDir, context.electronPlatformName)
     for (const filename of readdirSync(resourcesDir)) {
@@ -142,6 +158,10 @@ module.exports = {
     }
     if (context.electronPlatformName === 'darwin') {
       await signMacComputerUseHelper(join(resourcesDir, 'Orca Computer Use.app'), context.packager)
+      await signMacNotificationStatusHelper(
+        join(resourcesDir, '..', 'MacOS', 'orca-notification-status'),
+        context.packager
+      )
     }
   },
   win: {
@@ -173,7 +193,11 @@ module.exports = {
     artifactName: 'orca-windows-setup.${ext}',
     shortcutName: '${productName}',
     uninstallDisplayName: '${productName}',
-    createDesktopShortcut: 'always'
+    createDesktopShortcut: 'always',
+    // Why: on a real uninstall, stop and remove the relocated terminal daemon
+    // (which lives outside the install dir under LOCALAPPDATA by design). Guarded
+    // by ${isUpdated} inside so it never runs during an update's uninstallOldVersion.
+    include: resolve(__dirname, 'nsis', 'daemon-host-uninstall.nsh')
   },
   mac: {
     icon: 'resources/build/icon.icns',
@@ -229,6 +253,15 @@ module.exports = {
       },
       featureWallResources
     ],
+    // Why: the notification-status helper must execute from Contents/MacOS —
+    // on macOS 26 UNUserNotificationCenter aborts (bundleProxyForCurrentProcess
+    // is nil) for executables launched out of Contents/Resources (#7929).
+    extraFiles: [
+      {
+        from: 'native/notification-status-macos/.build/release/orca-notification-status',
+        to: 'MacOS/orca-notification-status'
+      }
+    ],
     target: [
       {
         target: 'dmg',
@@ -398,6 +431,37 @@ async function signMacComputerUseHelper(helperAppPath, packager) {
   })
 }
 
+async function signMacNotificationStatusHelper(helperPath, packager) {
+  if (!existsSync(helperPath)) {
+    if (isMacRelease) {
+      throw new Error(`Missing orca-notification-status helper at ${helperPath}`)
+    }
+    return
+  }
+  const codeSigningInfo =
+    isMacRelease && process.env.CSC_LINK && packager?.codeSigningInfo?.value
+      ? await packager.codeSigningInfo.value
+      : null
+  const identity =
+    process.env.CSC_NAME ??
+    findInstalledMacSigningIdentity(codeSigningInfo?.keychainFile) ??
+    (isMacRelease ? null : '-')
+  if (!identity) {
+    throw new Error('Missing signing identity for orca-notification-status helper')
+  }
+  // Why: macOS keys notification records to the code-signing identifier; the
+  // binary embeds the app's CFBundleIdentifier in __TEXT,__info_plist so this
+  // (and any later) `codesign --force` derives the correct identifier. Sign
+  // before the outer Orca.app is sealed, like the computer-use helper.
+  const args = ['--force', '--sign', identity]
+  if (isMacRelease) {
+    args.push('--options', 'runtime', '--timestamp')
+  }
+  args.push(helperPath)
+  execFileSync('codesign', args, { stdio: 'inherit' })
+  execFileSync('codesign', ['--verify', '--strict', helperPath], { stdio: 'inherit' })
+}
+
 function codesignArgs(identity, targetPath) {
   const args = ['--force', '--deep', '--sign', identity]
   if (isMacRelease) {

config/max-lines-baseline.txt

@@ -0,0 +1,359 @@
+# Files/globs currently allowed to exceed the oxlint `max-lines` budget.
+# This is a RATCHET: the list may only SHRINK. Do NOT add entries to get CI green —
+# split the oversized file instead (AGENTS.md → "Do Not Disable Max Lines").
+# Regenerate/prune: pnpm check:max-lines-ratchet --prune   (removes stale entries only)
+inline mobile/src/constants/marine-creatures.ts
+inline src/cli/handlers/automations.ts
+inline src/cli/handlers/orchestration.ts
+inline src/cli/help.ts
+inline src/cli/index.test.ts
+inline src/main/agent-hooks/server.test.ts
+inline src/main/agent-hooks/server.ts
+inline src/main/amp/hook-service.ts
+inline src/main/antigravity/hook-service.ts
+inline src/main/attribution/terminal-attribution.ts
+inline src/main/automations/external-manager.ts
+inline src/main/automations/hermes-cron-output.ts
+inline src/main/browser/agent-browser-bridge.test.ts
+inline src/main/browser/agent-browser-bridge.ts
+inline src/main/browser/browser-cookie-import.ts
+inline src/main/browser/browser-guest-ui.ts
+inline src/main/browser/browser-manager-grab.test.ts
+inline src/main/browser/browser-manager.test.ts
+inline src/main/browser/browser-manager.ts
+inline src/main/browser/browser-screencast-stream.ts
+inline src/main/browser/browser-session-registry.ts
+inline src/main/browser/cdp-bridge.ts
+inline src/main/browser/cdp-ws-proxy.ts
+inline src/main/browser/grab-guest-script.ts
+inline src/main/browser/snapshot-engine.ts
+inline src/main/claude-accounts/runtime-auth-service.test.ts
+inline src/main/claude-accounts/runtime-auth-service.ts
+inline src/main/claude-accounts/service.test.ts
+inline src/main/claude-accounts/service.ts
+inline src/main/claude-usage/scanner.ts
+inline src/main/claude-usage/store.ts
+inline src/main/cli/cli-installer.test.ts
+inline src/main/cli/cli-installer.ts
+inline src/main/cli/wsl-cli-installer.ts
+inline src/main/codex-accounts/runtime-home-service.test.ts
+inline src/main/codex-accounts/runtime-home-service.ts
+inline src/main/codex-accounts/service.test.ts
+inline src/main/codex-accounts/service.ts
+inline src/main/codex-usage/scanner.ts
+inline src/main/codex-usage/store.test.ts
+inline src/main/codex-usage/store.ts
+inline src/main/codex/config-toml-trust.test.ts
+inline src/main/codex/config-toml-trust.ts
+inline src/main/codex/hook-service.test.ts
+inline src/main/codex/hook-service.ts
+inline src/main/copilot/hook-service.ts
+inline src/main/daemon/client.ts
+inline src/main/daemon/daemon-health.ts
+inline src/main/daemon/daemon-init.test.ts
+inline src/main/daemon/daemon-init.ts
+inline src/main/daemon/daemon-pty-adapter.test.ts
+inline src/main/daemon/daemon-pty-adapter.ts
+inline src/main/daemon/daemon-server.ts
+inline src/main/daemon/pty-subprocess.test.ts
+inline src/main/daemon/pty-subprocess.ts
+inline src/main/daemon/session.ts
+inline src/main/daemon/shell-ready.ts
+inline src/main/git/remove-worktree.test.ts
+inline src/main/git/repo.ts
+inline src/main/git/runner.ts
+inline src/main/git/status.test.ts
+inline src/main/git/status.ts
+inline src/main/git/worktree.test.ts
+inline src/main/git/worktree.ts
+inline src/main/github/client.test.ts
+inline src/main/github/client.ts
+inline src/main/github/issues.ts
+inline src/main/github/pr-refresh-coordinator.test.ts
+inline src/main/github/pr-refresh-coordinator.ts
+inline src/main/github/project-view.ts
+inline src/main/github/project-view/mutations.ts
+inline src/main/github/work-item-details.ts
+inline src/main/gitlab/client-mr.test.ts
+inline src/main/gitlab/client.ts
+inline src/main/gitlab/issues.ts
+inline src/main/gitlab/work-item-details.ts
+inline src/main/hermes/hook-service.ts
+inline src/main/hooks.test.ts
+inline src/main/hooks.ts
+inline src/main/index.ts
+inline src/main/ipc/browser.ts
+inline src/main/ipc/crash-reporting.ts
+inline src/main/ipc/filesystem-auth.ts
+inline src/main/ipc/filesystem-mutations.ts
+inline src/main/ipc/filesystem-watcher.ts
+inline src/main/ipc/filesystem.test.ts
+inline src/main/ipc/filesystem.ts
+inline src/main/ipc/github.test.ts
+inline src/main/ipc/github.ts
+inline src/main/ipc/gitlab.ts
+inline src/main/ipc/linear.ts
+inline src/main/ipc/notifications.test.ts
+inline src/main/ipc/notifications.ts
+inline src/main/ipc/pet.ts
+inline src/main/ipc/preflight.test.ts
+inline src/main/ipc/pty.test.ts
+inline src/main/ipc/pty.ts
+inline src/main/ipc/remote-workspace.ts
+inline src/main/ipc/repos-remote.test.ts
+inline src/main/ipc/repos.ts
+inline src/main/ipc/runtime-environments.test.ts
+inline src/main/ipc/ssh.test.ts
+inline src/main/ipc/ssh.ts
+inline src/main/ipc/worktree-remote.ts
+inline src/main/ipc/worktrees.test.ts
+inline src/main/ipc/worktrees.ts
+inline src/main/jira/client.ts
+inline src/main/jira/issues.ts
+inline src/main/keybindings/keybinding-file.ts
+inline src/main/linear/client.ts
+inline src/main/linear/issues.ts
+inline src/main/linear/projects.ts
+inline src/main/memory/collector.ts
+inline src/main/opencode-usage/scanner.ts
+inline src/main/opencode-usage/store.ts
+inline src/main/opencode/hook-service.ts
+inline src/main/persistence.test.ts
+inline src/main/persistence.ts
+inline src/main/ports/advertised-url-watcher.ts
+inline src/main/ports/local-workspace-port-scanner.ts
+inline src/main/project-groups/nested-repo-discovery.ts
+inline src/main/providers/local-pty-provider.test.ts
+inline src/main/providers/local-pty-provider.ts
+inline src/main/providers/local-pty-shell-ready.test.ts
+inline src/main/providers/local-pty-shell-ready.ts
+inline src/main/providers/ssh-git-provider.test.ts
+inline src/main/providers/ssh-git-provider.ts
+inline src/main/rate-limits/claude-fetcher.test.ts
+inline src/main/rate-limits/claude-fetcher.ts
+inline src/main/rate-limits/claude-pty.ts
+inline src/main/rate-limits/codex-fetcher.ts
+inline src/main/rate-limits/service.test.ts
+inline src/main/rate-limits/service.ts
+inline src/main/runtime/orca-runtime-browser.ts
+inline src/main/runtime/orca-runtime-files.test.ts
+inline src/main/runtime/orca-runtime-files.ts
+inline src/main/runtime/orca-runtime-git.ts
+inline src/main/runtime/orca-runtime.test.ts
+inline src/main/runtime/orca-runtime.ts
+inline src/main/runtime/orchestration/coordinator.ts
+inline src/main/runtime/orchestration/db.ts
+inline src/main/runtime/rpc/methods/files.ts
+inline src/main/runtime/rpc/methods/git.ts
+inline src/main/runtime/rpc/methods/github.ts
+inline src/main/runtime/rpc/methods/orchestration.test.ts
+inline src/main/runtime/rpc/methods/orchestration.ts
+inline src/main/runtime/rpc/methods/terminal.ts
+inline src/main/runtime/rpc/terminal-multiplex.test.ts
+inline src/main/runtime/runtime-rpc.test.ts
+inline src/main/runtime/runtime-rpc.ts
+inline src/main/source-control/hosted-review-creation.ts
+inline src/main/speech/model-manager.ts
+inline src/main/speech/stt-service.ts
+inline src/main/ssh/ssh-channel-multiplexer.ts
+inline src/main/ssh/ssh-connection.test.ts
+inline src/main/ssh/ssh-connection.ts
+inline src/main/ssh/ssh-relay-deploy.ts
+inline src/main/ssh/ssh-relay-session.ts
+inline src/main/star-nag/service.test.ts
+inline src/main/text-generation/commit-message-text-generation.test.ts
+inline src/main/text-generation/commit-message-text-generation.ts
+inline src/main/updater.test.ts
+inline src/main/updater.ts
+inline src/main/window/attach-main-window-services.ts
+inline src/main/window/createMainWindow.test.ts
+inline src/main/window/createMainWindow.ts
+inline src/main/workspace-space-analysis.ts
+inline src/preload/api-types.ts
+inline src/preload/index.ts
+inline src/relay/agent-hook-server.ts
+inline src/relay/dispatcher.ts
+inline src/relay/external-automations-handler.ts
+inline src/relay/fs-handler.ts
+inline src/relay/git-handler.test.ts
+inline src/relay/git-handler.ts
+inline src/relay/pty-handler.test.ts
+inline src/relay/pty-handler.ts
+inline src/relay/relay.ts
+inline src/relay/workspace-space-scan.ts
+inline src/renderer/src/App.tsx
+inline src/renderer/src/components/GitHubItemDialog.tsx
+inline src/renderer/src/components/GitLabItemDialog.tsx
+inline src/renderer/src/components/JiraIssueWorkspace.tsx
+inline src/renderer/src/components/LinearIssueWorkspace.tsx
+inline src/renderer/src/components/LinearItemDrawer.tsx
+inline src/renderer/src/components/NewWorkspaceComposerCard.tsx
+inline src/renderer/src/components/PullRequestPage.tsx
+inline src/renderer/src/components/TaskPage.tsx
+inline src/renderer/src/components/Terminal.tsx
+inline src/renderer/src/components/UpdateCard.tsx
+inline src/renderer/src/components/WorktreeJumpPalette.tsx
+inline src/renderer/src/components/activity/ActivityPrototypePage.test.ts
+inline src/renderer/src/components/activity/ActivityPrototypePage.tsx
+inline src/renderer/src/components/automations/AutomationsPage.tsx
+inline src/renderer/src/components/browser-pane/BrowserPane.tsx
+inline src/renderer/src/components/diff-comments/useDiffCommentDecorator.tsx
+inline src/renderer/src/components/editor/CombinedDiffViewer.tsx
+inline src/renderer/src/components/editor/EditorContent.tsx
+inline src/renderer/src/components/editor/IpynbViewer.tsx
+inline src/renderer/src/components/editor/MarkdownPreview.tsx
+inline src/renderer/src/components/editor/MonacoEditor.tsx
+inline src/renderer/src/components/editor/editor-autosave-controller.ts
+inline src/renderer/src/components/editor/ipynb-parse.ts
+inline src/renderer/src/components/editor/useEditorPanelContentState.ts
+inline src/renderer/src/components/feature-wall/BrowserAnimatedVisual.tsx
+inline src/renderer/src/components/feature-wall/EditorAnimatedVisual.tsx
+inline src/renderer/src/components/feature-wall/WorkbenchAnimatedVisual.tsx
+inline src/renderer/src/components/floating-terminal/FloatingTerminalPanel.test.tsx
+inline src/renderer/src/components/floating-terminal/FloatingTerminalPanel.tsx
+inline src/renderer/src/components/github-project/ProjectCell.tsx
+inline src/renderer/src/components/github-project/ProjectPicker.tsx
+inline src/renderer/src/components/github-project/ProjectViewWrapper.tsx
+inline src/renderer/src/components/linear-project-view-surfaces.tsx
+inline src/renderer/src/components/new-workspace/SmartWorkspaceNameField.tsx
+inline src/renderer/src/components/onboarding/use-onboarding-flow.ts
+inline src/renderer/src/components/right-sidebar/ChecksPanel.tsx
+inline src/renderer/src/components/right-sidebar/FileExplorer.test.tsx
+inline src/renderer/src/components/right-sidebar/FileExplorer.tsx
+inline src/renderer/src/components/right-sidebar/FileExplorerRow.tsx
+inline src/renderer/src/components/right-sidebar/PortsPanel.tsx
+inline src/renderer/src/components/right-sidebar/SourceControl.tsx
+inline src/renderer/src/components/right-sidebar/checks-panel-content.tsx
+inline src/renderer/src/components/right-sidebar/index.tsx
+inline src/renderer/src/components/right-sidebar/source-control-dropdown-items.ts
+inline src/renderer/src/components/right-sidebar/useCreatePullRequestDialogFields.ts
+inline src/renderer/src/components/right-sidebar/useFileExplorerDragDrop.ts
+inline src/renderer/src/components/settings/AccountsPane.tsx
+inline src/renderer/src/components/settings/AgentsPane.tsx
+inline src/renderer/src/components/settings/RepositoryHooksSection.tsx
+inline src/renderer/src/components/settings/RuntimeEnvironmentsPane.tsx
+inline src/renderer/src/components/settings/Settings.tsx
+inline src/renderer/src/components/settings/SettingsFormControls.tsx
+inline src/renderer/src/components/sidebar/RemoteFileBrowser.tsx
+inline src/renderer/src/components/sidebar/WorkspaceKanbanDrawer.tsx
+inline src/renderer/src/components/sidebar/WorktreeCard.tsx
+inline src/renderer/src/components/sidebar/WorktreeContextMenu.tsx
+inline src/renderer/src/components/sidebar/WorktreeList.lineage-child-card.test.ts
+inline src/renderer/src/components/sidebar/WorktreeList.tsx
+inline src/renderer/src/components/sidebar/use-workspace-kanban-area-selection.ts
+inline src/renderer/src/components/sidebar/worktree-list-groups.test.ts
+inline src/renderer/src/components/sidebar/worktree-list-groups.ts
+inline src/renderer/src/components/stats/usage-overview-model.ts
+inline src/renderer/src/components/status-bar/ResourceUsageStatusSegment.tsx
+inline src/renderer/src/components/status-bar/StatusBar.tsx
+inline src/renderer/src/components/status-bar/WorkspaceSpaceManagerPanel.tsx
+inline src/renderer/src/components/tab-bar/TabBar.tsx
+inline src/renderer/src/components/tab-bar/TabBar.windows-shell-launch.test.ts
+inline src/renderer/src/components/tab-group/useTabDragSplit.ts
+inline src/renderer/src/components/tab-group/useTabGroupWorkspaceModel.ts
+inline src/renderer/src/components/task-page-cache-selectors.ts
+inline src/renderer/src/components/terminal-pane/TerminalPane.tsx
+inline src/renderer/src/components/terminal-pane/agent-completion-coordinator.test.ts
+inline src/renderer/src/components/terminal-pane/agent-completion-coordinator.ts
+inline src/renderer/src/components/terminal-pane/keyboard-handlers.ts
+inline src/renderer/src/components/terminal-pane/pty-connection.test.ts
+inline src/renderer/src/components/terminal-pane/pty-connection.ts
+inline src/renderer/src/components/terminal-pane/pty-transport.test.ts
+inline src/renderer/src/components/terminal-pane/pty-transport.ts
+inline src/renderer/src/components/terminal-pane/remote-runtime-pty-transport.test.ts
+inline src/renderer/src/components/terminal-pane/remote-runtime-pty-transport.ts
+inline src/renderer/src/components/terminal-pane/terminal-link-handlers.test.ts
+inline src/renderer/src/components/terminal-pane/use-terminal-pane-context-menu.ts
+inline src/renderer/src/components/terminal-pane/use-terminal-pane-global-effects.test.ts
+inline src/renderer/src/components/terminal-pane/use-terminal-pane-lifecycle.ts
+inline src/renderer/src/components/workspace-cleanup/WorkspaceCleanupDialog.tsx
+inline src/renderer/src/hooks/useAutomationDispatchEvents.ts
+inline src/renderer/src/hooks/useComposerState.ts
+inline src/renderer/src/hooks/useEditorExternalWatch.ts
+inline src/renderer/src/hooks/useIpcEvents.test.ts
+inline src/renderer/src/hooks/useIpcEvents.ts
+inline src/renderer/src/hooks/useIssueMetadata.ts
+inline src/renderer/src/hooks/useSettingsNavigationMetadata.ts
+inline src/renderer/src/lib/active-agent-note-send.test.ts
+inline src/renderer/src/lib/file-type-icons.ts
+inline src/renderer/src/lib/pane-manager/pane-manager.ts
+inline src/renderer/src/lib/pane-manager/pane-terminal-output-scheduler.test.ts
+inline src/renderer/src/lib/pane-manager/pane-terminal-output-scheduler.ts
+inline src/renderer/src/lib/pane-manager/pane-tree-ops.ts
+inline src/renderer/src/lib/terminal-links.ts
+inline src/renderer/src/lib/worktree-activation.test.ts
+inline src/renderer/src/lib/worktree-activation.ts
+inline src/renderer/src/runtime/remote-runtime-terminal-multiplexer.ts
+inline src/renderer/src/runtime/runtime-file-client.test.ts
+inline src/renderer/src/runtime/runtime-file-client.ts
+inline src/renderer/src/runtime/runtime-git-client.ts
+inline src/renderer/src/runtime/runtime-linear-client.ts
+inline src/renderer/src/runtime/sync-runtime-graph.test.ts
+inline src/renderer/src/runtime/sync-runtime-graph.ts
+inline src/renderer/src/runtime/web-runtime-session.test.ts
+inline src/renderer/src/runtime/web-runtime-session.ts
+inline src/renderer/src/runtime/web-session-tabs-sync.test.ts
+inline src/renderer/src/runtime/web-session-tabs-sync.ts
+inline src/renderer/src/store/slices/agent-status.test.ts
+inline src/renderer/src/store/slices/agent-status.ts
+inline src/renderer/src/store/slices/browser.test.ts
+inline src/renderer/src/store/slices/browser.ts
+inline src/renderer/src/store/slices/diffComments.ts
+inline src/renderer/src/store/slices/editor.test.ts
+inline src/renderer/src/store/slices/editor.ts
+inline src/renderer/src/store/slices/github.test.ts
+inline src/renderer/src/store/slices/github.ts
+inline src/renderer/src/store/slices/hosted-review.ts
+inline src/renderer/src/store/slices/jira.ts
+inline src/renderer/src/store/slices/linear.test.ts
+inline src/renderer/src/store/slices/linear.ts
+inline src/renderer/src/store/slices/repos.ts
+inline src/renderer/src/store/slices/store-cascades.test.ts
+inline src/renderer/src/store/slices/store-session-cascades.test.ts
+inline src/renderer/src/store/slices/tabs.test.ts
+inline src/renderer/src/store/slices/tabs.ts
+inline src/renderer/src/store/slices/terminals.ts
+inline src/renderer/src/store/slices/ui.test.ts
+inline src/renderer/src/store/slices/ui.ts
+inline src/renderer/src/store/slices/workspace-cleanup.ts
+inline src/renderer/src/store/slices/worktrees.test.ts
+inline src/renderer/src/store/slices/worktrees.ts
+inline src/renderer/src/web/web-preload-api.test.ts
+inline src/renderer/src/web/web-preload-api.ts
+inline src/renderer/src/web/web-runtime-client.ts
+inline src/shared/agent-hook-listener.test.ts
+inline src/shared/agent-hook-listener.ts
+inline src/shared/automation-schedules.ts
+inline src/shared/commit-message-agent-spec.ts
+inline src/shared/constants.ts
+inline src/shared/github-project-types.ts
+inline src/shared/keybindings.test.ts
+inline src/shared/keybindings.ts
+inline src/shared/marine-creatures.ts
+inline src/shared/remote-runtime-client.ts
+inline src/shared/runtime-types.ts
+inline src/shared/source-control-ai.ts
+inline src/shared/telemetry-events.ts
+inline src/shared/text-search.ts
+inline src/shared/types.ts
+inline tests/e2e/helpers/terminal.ts
+inline tests/e2e/terminal-panes.spec.ts
+mobile-config app/h/*/files/*.tsx
+mobile-config app/h/*/index.tsx
+mobile-config app/h/*/session/*.tsx
+mobile-config app/h/*/source-control/*.tsx
+mobile-config app/h/*/tasks.tsx
+mobile-config app/index.tsx
+mobile-config app/pair-scan.tsx
+mobile-config app/terminal-settings.tsx
+mobile-config app/troubleshoot.tsx
+mobile-config scripts/mock-server.ts
+mobile-config scripts/repro-terminal-colors.ts
+mobile-config scripts/repro-worktree-startup-stream.ts
+mobile-config src/browser/MobileBrowserPane.tsx
+mobile-config src/components/CustomKeyModal.tsx
+mobile-config src/components/NewWorktreeModal.tsx
+mobile-config src/components/mobile-rich-markdown-editor-html.ts
+mobile-config src/terminal/terminal-accessory-keys.ts
+mobile-config src/terminal/terminal-webview-html.ts
+mobile-config src/transport/rpc-client.ts

config/nsis/daemon-host-uninstall.nsh

@@ -0,0 +1,23 @@
+; Clean up the relocated terminal daemon on a REAL uninstall.
+;
+; Why: the daemon host is deliberately copied to a distinct image name
+; (orca-terminal-daemon.exe) under %LOCALAPPDATA%\Orca\daemon-host so that app
+; UPDATES cannot kill it — that relocation is what keeps terminals alive across
+; updates. The same design means a normal uninstall's process sweep and file
+; removal both miss it, leaving an orphaned daemon plus its runtime copy behind.
+;
+; The ${isUpdated} guard is essential: electron-builder runs this uninstaller as
+; part of uninstallOldVersion on EVERY update, and killing the daemon there would
+; defeat the whole feature. Only clean up on a genuine uninstall.
+;
+; The image name and the LOCALAPPDATA folder name must stay in sync with
+; DAEMON_HOST_EXE_NAME and LOCAL_HOST_ROOT_NAME in
+; src/main/daemon/daemon-host-relocation.ts.
+!macro customUnInstall
+  ${ifNot} ${isUpdated}
+    nsExec::Exec 'taskkill /F /IM orca-terminal-daemon.exe'
+    ; Give the OS a moment to release the image lock before removing the tree.
+    Sleep 500
+    RMDir /r "$LOCALAPPDATA\Orca\daemon-host"
+  ${endIf}
+!macroend

config/scripts/build-native-for-platform.mjs

@@ -8,6 +8,8 @@ if (process.platform !== 'darwin') {
 }
 
 runPnpmScript('build:computer-macos')
+runPnpmScript('build:notification-status-macos')
+process.exit(0)
 
 function runPnpmScript(scriptName) {
   const npmExecPath = process.env.npm_execpath
@@ -22,5 +24,7 @@ function runPnpmScript(scriptName) {
   if (result.signal) {
     process.kill(process.pid, result.signal)
   }
-  process.exit(result.status ?? (result.error ? 1 : 0))
+  if (result.status !== 0 || result.error) {
+    process.exit(result.status ?? 1)
+  }
 }

config/scripts/build-notification-status-macos.mjs

@@ -0,0 +1,97 @@
+#!/usr/bin/env node
+// Builds the orca-notification-status helper binary.
+//
+// The helper reads UNUserNotificationCenter settings for the app it ships
+// inside (see native/notification-status-macos/main.swift). The target
+// CFBundleIdentifier is embedded as a __TEXT,__info_plist section so every
+// later `codesign --force` pass (electron-builder's signing, the dev runner's
+// ad-hoc deep sign) derives the correct code identifier automatically —
+// macOS keys notification records to that identifier.
+import { execFileSync } from 'node:child_process'
+import { mkdirSync, rmSync, writeFileSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import path from 'node:path'
+
+const repoRoot = path.resolve(import.meta.dirname, '../..')
+const sourcePath = path.join(repoRoot, 'native', 'notification-status-macos', 'main.swift')
+const defaultOutputPath = path.join(
+  repoRoot,
+  'native',
+  'notification-status-macos',
+  '.build',
+  'release',
+  'orca-notification-status'
+)
+
+if (process.platform !== 'darwin') {
+  process.exit(0)
+}
+
+const args = process.argv.slice(2)
+const bundleId = readArg('--bundle-id') ?? 'com.stablyai.orca'
+const outputPath = readArg('--output') ?? defaultOutputPath
+// Why: dev launches only need the host architecture; release builds ship a
+// universal binary matching the app's x64 + arm64 targets.
+const singleArch = args.includes('--single-arch')
+
+const workDir = path.join(tmpdir(), `orca-notification-status-${process.pid}`)
+mkdirSync(workDir, { recursive: true })
+try {
+  const plistPath = path.join(workDir, 'Info.plist')
+  writeFileSync(plistPath, embeddedInfoPlist(bundleId), 'utf8')
+  const triples = singleArch
+    ? [process.arch === 'arm64' ? 'arm64-apple-macosx' : 'x86_64-apple-macosx']
+    : ['arm64-apple-macosx', 'x86_64-apple-macosx']
+  const builtBinaries = triples.map((triple) => {
+    const output = path.join(workDir, `orca-notification-status-${triple}`)
+    execFileSync(
+      'swiftc',
+      [
+        '-O',
+        sourcePath,
+        '-target',
+        triple.replace('-apple-macosx', '-apple-macosx11.0'),
+        '-o',
+        output,
+        '-Xlinker',
+        '-sectcreate',
+        '-Xlinker',
+        '__TEXT',
+        '-Xlinker',
+        '__info_plist',
+        '-Xlinker',
+        plistPath
+      ],
+      { stdio: 'inherit' }
+    )
+    return output
+  })
+  mkdirSync(path.dirname(outputPath), { recursive: true })
+  if (builtBinaries.length === 1) {
+    execFileSync('cp', [builtBinaries[0], outputPath])
+  } else {
+    execFileSync('lipo', ['-create', ...builtBinaries, '-output', outputPath])
+  }
+  execFileSync('chmod', ['755', outputPath])
+} finally {
+  rmSync(workDir, { recursive: true, force: true })
+}
+
+function readArg(name) {
+  const index = args.indexOf(name)
+  return index >= 0 ? args[index + 1] : undefined
+}
+
+function embeddedInfoPlist(identifier) {
+  return `<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+<dict>
+  <key>CFBundleIdentifier</key>
+  <string>${identifier}</string>
+  <key>CFBundleName</key>
+  <string>orca-notification-status</string>
+</dict>
+</plist>
+`
+}

config/scripts/check-max-lines-ratchet.mjs

@@ -0,0 +1,254 @@
+import { execFileSync } from 'node:child_process'
+import fs from 'node:fs'
+import path from 'node:path'
+import process from 'node:process'
+import { pathToFileURL } from 'node:url'
+
+// Ratchet gate for the oxlint `max-lines` rule.
+//
+// oxlint already fails any file that exceeds max-lines WITHOUT a suppression, so
+// the only way a file grows past the budget is by adding an `eslint/oxlint-disable
+// max-lines` comment or a per-file `max-lines` bump in mobile/.oxlintrc.json. This
+// check freezes the set of files currently allowed to do that (the baseline) and
+// fails CI when a NEW bypass appears — the existing over-limit files are
+// grandfathered; new ones must split instead. The baseline may only shrink.
+
+const BASELINE_PATH = 'config/max-lines-baseline.txt'
+const MOBILE_CONFIG_PATH = 'mobile/.oxlintrc.json'
+// These two files legitimately contain the directive text as data (regex, fixtures),
+// so scanning them would self-flag. The ratchet does not police itself.
+const SELF_FILES = new Set([
+  'config/scripts/check-max-lines-ratchet.mjs',
+  'config/scripts/check-max-lines-ratchet.test.mjs'
+])
+
+// Default max-lines budgets from .oxlintrc.json (counted lines).
+export function defaultLimitForPath(p) {
+  if (/\.(test|spec)\.(ts|tsx)$/.test(p)) {
+    return 800
+  }
+  if (p.endsWith('.tsx')) {
+    return 400
+  }
+  if (p.endsWith('.mjs')) {
+    return 600
+  }
+  return 300
+}
+
+// True if the source contains an eslint/oxlint disable directive listing `max-lines`
+// (block or line, bare or compound, with or without a `-- Why:` reason).
+export function hasMaxLinesDisable(sourceText) {
+  const re = /(?:eslint|oxlint)-disable(?:-next-line|-line)?\b([^\n]*)/g
+  let m
+  while ((m = re.exec(sourceText)) !== null) {
+    let rules = m[1]
+    rules = rules.split('--')[0] // strip the reason
+    const close = rules.indexOf('*/')
+    if (close !== -1) {
+      rules = rules.slice(0, close) // strip block-comment tail
+    }
+    if (/\bmax-lines\b/.test(rules)) {
+      return true
+    }
+  }
+  return false
+}
+
+// Per-file `max-lines` bumps in mobile/.oxlintrc.json whose `max` exceeds the
+// default for that glob (a lower `max` is stricter, not a bypass).
+export function collectMobileBumps(configText) {
+  const cfg = JSON.parse(configText)
+  const bumps = []
+  for (const override of cfg.overrides ?? []) {
+    const rule = override.rules?.['max-lines']
+    if (!Array.isArray(rule) || typeof rule[1]?.max !== 'number') {
+      continue
+    }
+    for (const glob of override.files ?? []) {
+      if (rule[1].max > defaultLimitForPath(glob)) {
+        bumps.push(`mobile-config ${glob}`)
+      }
+    }
+  }
+  return bumps
+}
+
+export function parseBaseline(text) {
+  return new Set(
+    text
+      .split('\n')
+      .map((l) => l.trim())
+      .filter((l) => l && !l.startsWith('#'))
+  )
+}
+
+export function diffBaseline(current, baseline) {
+  const cur = new Set(current)
+  const base = baseline instanceof Set ? baseline : new Set(baseline)
+  const added = [...cur].filter((e) => !base.has(e)).sort()
+  const stale = [...base].filter((e) => !cur.has(e)).sort()
+  return { added, stale }
+}
+
+// Collect every current suppression entry from the tracked tree.
+export function collectCurrentSuppressions(root = process.cwd()) {
+  const tracked = execFileSync('git', ['ls-files', '*.ts', '*.tsx', '*.mjs'], {
+    cwd: root,
+    encoding: 'utf8',
+    maxBuffer: 64 * 1024 * 1024
+  })
+    .split('\n')
+    .filter(Boolean)
+    .filter((f) => !SELF_FILES.has(f))
+
+  const entries = []
+  for (const rel of tracked) {
+    let src
+    try {
+      src = fs.readFileSync(path.join(root, rel), 'utf8')
+    } catch {
+      continue
+    }
+    if (hasMaxLinesDisable(src)) {
+      entries.push(`inline ${rel}`)
+    }
+  }
+
+  const mobileCfgPath = path.join(root, MOBILE_CONFIG_PATH)
+  if (fs.existsSync(mobileCfgPath)) {
+    entries.push(...collectMobileBumps(fs.readFileSync(mobileCfgPath, 'utf8')))
+  }
+
+  return entries.sort()
+}
+
+function printAddedFailure(added) {
+  for (const entry of added) {
+    console.error(`::error::New max-lines bypass not allowed: ${entry}`)
+  }
+  console.error('')
+  console.error('╭────────────────────────────────────────────────────────────────────────────╮')
+  console.error('│  ❌  max-lines ratchet failed — a NEW file is trying to exceed the line cap.  │')
+  console.error('╰────────────────────────────────────────────────────────────────────────────╯')
+  console.error('')
+  console.error(`  ${added.length} file(s)/glob(s) newly bypass the oxlint \`max-lines\` rule:`)
+  console.error('')
+  for (const entry of added) {
+    const [kind, ...rest] = entry.split(' ')
+    const target = rest.join(' ')
+    const how =
+      kind === 'inline'
+        ? 'added an eslint/oxlint-disable max-lines comment'
+        : 'added a per-file max-lines bump in mobile/.oxlintrc.json'
+    console.error(`    • ${target}\n        ↳ ${how}`)
+  }
+  console.error('')
+  console.error('  Orca caps file size (300 .ts / 400 .tsx / 600 .mjs / 800 test — non-blank,')
+  console.error(
+    '  non-comment lines). Existing oversized files are grandfathered; NEW ones are not.'
+  )
+  console.error('')
+  console.error('  ✅  Fix it: SPLIT the file into focused modules — do NOT suppress the rule.')
+  console.error('      See AGENTS.md → "Lint Rules: Do Not Disable Max Lines".')
+  console.error('')
+  console.error('  (If you are intentionally, with reviewer sign-off, adding an unavoidable')
+  console.error(`   exception, add the exact line(s) above to ${BASELINE_PATH}.)`)
+  console.error('')
+}
+
+function printStaleFailure(stale) {
+  for (const entry of stale) {
+    console.error(`::error::Stale max-lines baseline entry (prune it): ${entry}`)
+  }
+  console.error('')
+  console.error('╭────────────────────────────────────────────────────────────────────────────╮')
+  console.error('│  ⚠️  max-lines baseline is out of date — nice work removing a bypass!         │')
+  console.error('╰────────────────────────────────────────────────────────────────────────────╯')
+  console.error('')
+  console.error(`  ${stale.length} baseline entr(y/ies) no longer have a max-lines suppression.`)
+  console.error(
+    '  The baseline may only shrink, so these must be removed to keep re-adding blocked:'
+  )
+  console.error('')
+  for (const entry of stale) {
+    console.error(`    • ${entry}`)
+  }
+  console.error('')
+  console.error(`  ✅  Fix it (one command):  pnpm check:max-lines-ratchet --prune`)
+  console.error('')
+}
+
+export function main(root = process.cwd()) {
+  const baselineFile = path.join(root, BASELINE_PATH)
+  if (!fs.existsSync(baselineFile)) {
+    console.error(
+      `::error::Missing ${BASELINE_PATH}. Generate it with: node config/scripts/check-max-lines-ratchet.mjs --init`
+    )
+    return 1
+  }
+  const baseline = parseBaseline(fs.readFileSync(baselineFile, 'utf8'))
+  const current = collectCurrentSuppressions(root)
+  const { added, stale } = diffBaseline(current, baseline)
+
+  if (added.length > 0) {
+    printAddedFailure(added)
+    if (stale.length > 0) {
+      console.error(
+        `  (Also: ${stale.length} stale baseline entr(y/ies) can be pruned — see below.)`
+      )
+      printStaleFailure(stale)
+    }
+    return 1
+  }
+  if (stale.length > 0) {
+    printStaleFailure(stale)
+    return 1
+  }
+  console.log(
+    `max-lines ratchet OK — ${current.length} grandfathered suppression(s), no new bypasses.`
+  )
+  return 0
+}
+
+function writeBaseline(root, entries) {
+  const header = [
+    '# Files/globs currently allowed to exceed the oxlint `max-lines` budget.',
+    '# This is a RATCHET: the list may only SHRINK. Do NOT add entries to get CI green —',
+    '# split the oversized file instead (AGENTS.md → "Do Not Disable Max Lines").',
+    '# Regenerate/prune: pnpm check:max-lines-ratchet --prune   (removes stale entries only)',
+    ''
+  ].join('\n')
+  fs.writeFileSync(path.join(root, BASELINE_PATH), `${header}${entries.join('\n')}\n`)
+}
+
+if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) {
+  const root = process.cwd()
+  const arg = process.argv[2]
+  if (arg === '--init') {
+    // One-time bootstrap: capture the current suppression set as the baseline.
+    const entries = collectCurrentSuppressions(root)
+    writeBaseline(root, entries)
+    console.log(`Wrote ${BASELINE_PATH} with ${entries.length} entries.`)
+    process.exit(0)
+  }
+  if (arg === '--prune') {
+    // Remove baseline entries whose suppression is gone (shrink only; never adds).
+    const current = new Set(collectCurrentSuppressions(root))
+    const baseline = parseBaseline(fs.readFileSync(path.join(root, BASELINE_PATH), 'utf8'))
+    const kept = [...baseline].filter((e) => current.has(e)).sort()
+    const newlyAdded = [...current].filter((e) => !baseline.has(e))
+    writeBaseline(root, kept)
+    console.log(
+      `Pruned baseline to ${kept.length} entries (removed ${baseline.size - kept.length}).`
+    )
+    if (newlyAdded.length > 0) {
+      console.error(
+        `::error::--prune does not add entries; ${newlyAdded.length} new bypass(es) remain — split those files.`
+      )
+      process.exit(1)
+    }
+    process.exit(0)
+  }
+  process.exit(main(root))
+}

config/scripts/check-max-lines-ratchet.test.mjs

@@ -0,0 +1,115 @@
+import { describe, expect, it } from 'vitest'
+
+import {
+  collectMobileBumps,
+  defaultLimitForPath,
+  diffBaseline,
+  hasMaxLinesDisable,
+  parseBaseline
+} from './check-max-lines-ratchet.mjs'
+
+describe('hasMaxLinesDisable', () => {
+  it('detects a bare block disable', () => {
+    expect(hasMaxLinesDisable('/* eslint-disable max-lines */\nexport const a = 1\n')).toBe(true)
+  })
+
+  it('detects the oxlint spelling', () => {
+    expect(hasMaxLinesDisable('/* oxlint-disable max-lines */\n')).toBe(true)
+  })
+
+  it('detects a disable with a -- Why reason', () => {
+    expect(hasMaxLinesDisable('/* eslint-disable max-lines -- Why: one owner. */\n')).toBe(true)
+  })
+
+  it('detects a multi-line block where the reason wraps', () => {
+    const src =
+      '/* eslint-disable max-lines -- Why: this contract is\n * intentionally centralized. */\nimport x from "y"\n'
+    expect(hasMaxLinesDisable(src)).toBe(true)
+  })
+
+  it('detects max-lines inside a compound rule list', () => {
+    expect(hasMaxLinesDisable('/* eslint-disable no-control-regex, max-lines -- Why: x */\n')).toBe(
+      true
+    )
+    expect(hasMaxLinesDisable('/* eslint-disable max-lines, no-control-regex */\n')).toBe(true)
+  })
+
+  it('detects a line-scoped disable', () => {
+    expect(hasMaxLinesDisable('const a = 1 // eslint-disable-line max-lines\n')).toBe(true)
+  })
+
+  it('ignores a disable for an unrelated rule', () => {
+    expect(hasMaxLinesDisable('/* eslint-disable no-console */\n')).toBe(false)
+  })
+
+  it('does not treat "max-lines" appearing only in the reason text as a suppression', () => {
+    // max-lines is after the `--`, so it is prose, not a suppressed rule.
+    expect(
+      hasMaxLinesDisable('/* eslint-disable no-console -- we could hit max-lines later */\n')
+    ).toBe(false)
+  })
+
+  it('returns false for ordinary source', () => {
+    expect(hasMaxLinesDisable('export function f() {\n  return 42\n}\n')).toBe(false)
+  })
+})
+
+describe('defaultLimitForPath', () => {
+  it('uses 800 for tests, 400 for tsx, 600 for mjs, 300 otherwise', () => {
+    expect(defaultLimitForPath('a/b.test.ts')).toBe(800)
+    expect(defaultLimitForPath('a/b.spec.tsx')).toBe(800)
+    expect(defaultLimitForPath('a/b.tsx')).toBe(400)
+    expect(defaultLimitForPath('a/b.mjs')).toBe(600)
+    expect(defaultLimitForPath('a/b.ts')).toBe(300)
+  })
+})
+
+describe('collectMobileBumps', () => {
+  it('captures only overrides whose max exceeds the default for the glob', () => {
+    const cfg = JSON.stringify({
+      overrides: [
+        { files: ['app/h/*/tasks.tsx'], rules: { 'max-lines': ['error', { max: 14682 }] } }, // bump (>400)
+        {
+          files: ['src/terminal/TerminalWebView.tsx'],
+          rules: { 'max-lines': ['error', { max: 379 }] }
+        }, // stricter (<400), skip
+        { files: ['scripts/mock-server.ts'], rules: { 'max-lines': ['error', { max: 407 }] } } // bump (>300)
+      ]
+    })
+    expect(collectMobileBumps(cfg)).toEqual([
+      'mobile-config app/h/*/tasks.tsx',
+      'mobile-config scripts/mock-server.ts'
+    ])
+  })
+
+  it('ignores overrides without a max-lines rule', () => {
+    const cfg = JSON.stringify({
+      overrides: [{ files: ['a.tsx'], rules: { 'no-console': 'off' } }]
+    })
+    expect(collectMobileBumps(cfg)).toEqual([])
+  })
+})
+
+describe('parseBaseline', () => {
+  it('drops comments and blank lines', () => {
+    const b = parseBaseline('# header\n\ninline a.ts\nmobile-config x/*.tsx\n')
+    expect(b).toEqual(new Set(['inline a.ts', 'mobile-config x/*.tsx']))
+  })
+})
+
+describe('diffBaseline', () => {
+  it('reports added and stale entries', () => {
+    const { added, stale } = diffBaseline(
+      ['inline b.ts', 'inline c.ts'],
+      new Set(['inline a.ts', 'inline b.ts'])
+    )
+    expect(added).toEqual(['inline c.ts']) // new bypass
+    expect(stale).toEqual(['inline a.ts']) // suppression removed
+  })
+
+  it('is clean when current matches baseline', () => {
+    const { added, stale } = diffBaseline(['inline a.ts'], new Set(['inline a.ts']))
+    expect(added).toEqual([])
+    expect(stale).toEqual([])
+  })
+})

config/scripts/check-reliability-gates.mjs

@@ -0,0 +1,431 @@
+import fs from 'node:fs/promises'
+import path from 'node:path'
+import { pathToFileURL } from 'node:url'
+import process from 'node:process'
+
+import { parse, printParseErrorCode } from 'jsonc-parser'
+
+const MANIFEST_PATH = path.join('config', 'reliability-gates.jsonc')
+const RED_GREEN_STATUSES = new Set(['missing', 'partial', 'complete', 'not-required'])
+const FLAKE_STATUSES = new Set(['not-started', 'unknown', 'soaking', 'stable', 'flaky'])
+const PROTECTION_LEVELS = new Set(['none', 'partial', 'active'])
+const EVIDENCE_RUN_RESULTS = new Set(['passed', 'failed', 'skipped'])
+const EVIDENCE_RUNNERS = new Set(['local', 'ci', 'soak', 'manual'])
+
+function isRecord(value) {
+  return value !== null && typeof value === 'object' && !Array.isArray(value)
+}
+
+function isNonEmptyString(value) {
+  return typeof value === 'string' && value.trim().length > 0
+}
+
+function hasNonEmptyStringArray(value) {
+  return Array.isArray(value) && value.length > 0 && value.every(isNonEmptyString)
+}
+
+function requireNonEmptyString(gate, field, failures) {
+  if (!isNonEmptyString(gate[field])) {
+    failures.push(`${gate.id ?? '<unknown>'}: ${field} must be a non-empty string`)
+  }
+}
+
+function requireStringArray(gate, field, failures) {
+  if (!hasNonEmptyStringArray(gate[field])) {
+    failures.push(`${gate.id ?? '<unknown>'}: ${field} must be a non-empty string array`)
+  }
+}
+
+function requireStringArrayAllowEmpty(gate, field, failures) {
+  if (!Array.isArray(gate[field]) || !gate[field].every(isNonEmptyString)) {
+    failures.push(`${gate.id ?? '<unknown>'}: ${field} must be an array of strings`)
+  }
+}
+
+function requireNonNegativeNumber(record, field, owner, failures) {
+  if (!Number.isFinite(record[field]) || record[field] < 0) {
+    failures.push(`${owner}.${field} must be a non-negative number`)
+  }
+}
+
+function validatePolicy(manifest, failures) {
+  if (!isRecord(manifest.policy)) {
+    failures.push('policy must be an object')
+    return new Set()
+  }
+  if (!hasNonEmptyStringArray(manifest.policy.maturityLevels)) {
+    failures.push('policy.maturityLevels must be a non-empty string array')
+  }
+  if (!isRecord(manifest.policy.blockingPromotion)) {
+    failures.push('policy.blockingPromotion must be an object')
+  } else {
+    for (const field of ['minimumSoakRuns', 'minimumSoakDays', 'maximumUnexplainedFlakes']) {
+      requireNonNegativeNumber(
+        manifest.policy.blockingPromotion,
+        field,
+        'policy.blockingPromotion',
+        failures
+      )
+    }
+  }
+  return new Set(
+    Array.isArray(manifest.policy.maturityLevels)
+      ? manifest.policy.maturityLevels.filter(isNonEmptyString)
+      : []
+  )
+}
+
+function validateRuntimeBudget(gate, failures) {
+  if (!isRecord(gate.runtimeBudget)) {
+    failures.push(`${gate.id}: runtimeBudget must be an object`)
+    return
+  }
+  if (!Number.isFinite(gate.runtimeBudget.p95Seconds) || gate.runtimeBudget.p95Seconds <= 0) {
+    failures.push(`${gate.id}: runtimeBudget.p95Seconds must be a positive number`)
+  }
+  if (!isNonEmptyString(gate.runtimeBudget.scope)) {
+    failures.push(`${gate.id}: runtimeBudget.scope must be a non-empty string`)
+  }
+}
+
+function validateEvidence(gate, field, allowedStatuses, failures) {
+  const evidence = gate[field]
+  if (!isRecord(evidence)) {
+    failures.push(`${gate.id}: ${field} must be an object`)
+    return
+  }
+  if (!allowedStatuses.has(evidence.status)) {
+    failures.push(`${gate.id}: ${field}.status is invalid`)
+  }
+  if (!isNonEmptyString(evidence.evidence)) {
+    failures.push(`${gate.id}: ${field}.evidence must be a non-empty string`)
+  }
+}
+
+function validatePerformanceBudget(gate, failures) {
+  if (!isRecord(gate.performanceBudget)) {
+    failures.push(`${gate.id}: performanceBudget must be an object`)
+    return
+  }
+  if (typeof gate.performanceBudget.required !== 'boolean') {
+    failures.push(`${gate.id}: performanceBudget.required must be boolean`)
+  }
+  if (!isNonEmptyString(gate.performanceBudget.evidence)) {
+    failures.push(`${gate.id}: performanceBudget.evidence must be a non-empty string`)
+  }
+}
+
+function commandUsesBrittleTestSelector(command) {
+  return /(?:^|\s)(?:-t|--testNamePattern|--grep)(?:=|\s)/.test(command)
+}
+
+function validateCommandCoverage(gate, failures) {
+  if (!hasNonEmptyStringArray(gate.commands) || !hasNonEmptyStringArray(gate.testFiles)) {
+    return
+  }
+  for (const testFile of gate.testFiles) {
+    if (!gate.commands.some((command) => command.includes(testFile))) {
+      failures.push(`${gate.id}: test file is not referenced by any gate command: ${testFile}`)
+    }
+  }
+}
+
+function validateProtection(gate, failures) {
+  if (!PROTECTION_LEVELS.has(gate.protection)) {
+    failures.push(`${gate.id}: protection must be one of none, partial, or active`)
+    return
+  }
+  const hasCommands = hasNonEmptyStringArray(gate.commands)
+  const hasTestFiles = hasNonEmptyStringArray(gate.testFiles)
+  if (gate.protection === 'none' && (hasCommands || hasTestFiles)) {
+    failures.push(`${gate.id}: protection none gates must not declare commands or testFiles`)
+  }
+  if (gate.protection === 'partial' && (!hasCommands || !hasTestFiles)) {
+    failures.push(`${gate.id}: protection partial gates must declare commands and testFiles`)
+  }
+  if (gate.protection === 'active') {
+    if (gate.maturity !== 'blocking') {
+      failures.push(`${gate.id}: protection active is reserved for blocking gates`)
+    }
+    if (!hasCommands || !hasTestFiles) {
+      failures.push(`${gate.id}: protection active gates must declare commands and testFiles`)
+    }
+    if (gate.flakeHistory?.status !== 'stable') {
+      failures.push(`${gate.id}: protection active gates must have stable flakeHistory`)
+    }
+    if (!hasCompleteRedGreenEvidence(gate)) {
+      failures.push(`${gate.id}: protection active gates must have complete red/green evidence`)
+    }
+  }
+  if (!hasCommands && gate.protection !== 'none') {
+    failures.push(`${gate.id}: commandless gates must declare protection none`)
+  }
+}
+
+function isIsoDate(value) {
+  return typeof value === 'string' && /^\d{4}-\d{2}-\d{2}$/.test(value)
+}
+
+// Why: cross-field validators run even after an earlier type check flagged a
+// malformed field, so coerce to an array before `.includes` to report the error
+// instead of throwing on hand-edited manifests.
+function asArray(value) {
+  return Array.isArray(value) ? value : []
+}
+
+function hasCompleteRedGreenEvidence(gate) {
+  return ['complete', 'not-required'].includes(gate.redGreenEvidence?.status)
+}
+
+function validateEvidenceRun(gate, run, index, failures) {
+  const owner = `${gate.id}: evidenceRuns[${index}]`
+  if (!isRecord(run)) {
+    failures.push(`${owner} must be an object`)
+    return
+  }
+  if (!isIsoDate(run.date)) {
+    failures.push(`${owner}.date must be YYYY-MM-DD`)
+  }
+  if (!EVIDENCE_RUNNERS.has(run.runner)) {
+    failures.push(`${owner}.runner must be one of local, ci, soak, or manual`)
+  }
+  if (!isNonEmptyString(run.platform)) {
+    failures.push(`${owner}.platform must be a non-empty string`)
+  }
+  if (!EVIDENCE_RUN_RESULTS.has(run.result)) {
+    failures.push(`${owner}.result must be one of passed, failed, or skipped`)
+  }
+  if (!isNonEmptyString(run.command)) {
+    failures.push(`${owner}.command must be a non-empty string`)
+  } else if (!asArray(gate.commands).includes(run.command)) {
+    failures.push(`${owner}.command must match one of the gate commands`)
+  }
+  if (!Number.isFinite(run.durationSeconds) || run.durationSeconds < 0) {
+    failures.push(`${owner}.durationSeconds must be a non-negative number`)
+  }
+  if (!isNonEmptyString(run.summary)) {
+    failures.push(`${owner}.summary must be a non-empty string`)
+  }
+}
+
+function validateEvidenceRuns(gate, failures) {
+  if (!Array.isArray(gate.evidenceRuns)) {
+    failures.push(`${gate.id}: evidenceRuns must be an array`)
+    return
+  }
+  gate.evidenceRuns.forEach((run, index) => validateEvidenceRun(gate, run, index, failures))
+  if (
+    ['partial', 'active'].includes(gate.protection) &&
+    !gate.evidenceRuns.some((run) => isRecord(run) && run.result === 'passed')
+  ) {
+    failures.push(`${gate.id}: protection ${gate.protection} gates need a passed evidence run`)
+  }
+  if (gate.protection === 'none' && gate.evidenceRuns.length > 0) {
+    failures.push(`${gate.id}: protection none gates must not declare evidenceRuns`)
+  }
+}
+
+function validateAssertionRef(gate, ref, index, failures) {
+  const owner = `${gate.id}: assertionRefs[${index}]`
+  if (!isRecord(ref)) {
+    failures.push(`${owner} must be an object`)
+    return
+  }
+  if (!isNonEmptyString(ref.file)) {
+    failures.push(`${owner}.file must be a non-empty string`)
+  } else if (!asArray(gate.testFiles).includes(ref.file)) {
+    failures.push(`${owner}.file must be one of the gate testFiles`)
+  }
+  if (!hasNonEmptyStringArray(ref.assertions)) {
+    failures.push(`${owner}.assertions must be a non-empty string array`)
+  }
+}
+
+function validateAssertionRefs(gate, failures) {
+  if (!Array.isArray(gate.assertionRefs)) {
+    failures.push(`${gate.id}: assertionRefs must be an array`)
+    return
+  }
+  gate.assertionRefs.forEach((ref, index) => validateAssertionRef(gate, ref, index, failures))
+  if (['partial', 'active'].includes(gate.protection) && gate.assertionRefs.length === 0) {
+    failures.push(`${gate.id}: protection ${gate.protection} gates need assertionRefs`)
+  }
+  if (gate.protection === 'none' && gate.assertionRefs.length > 0) {
+    failures.push(`${gate.id}: protection none gates must not declare assertionRefs`)
+  }
+}
+
+function validateCoveredScope(gate, failures) {
+  requireStringArrayAllowEmpty(gate, 'coveredPlatforms', failures)
+  requireStringArrayAllowEmpty(gate, 'coveredProviders', failures)
+  requireNonEmptyString(gate, 'coverageNotes', failures)
+
+  if (!Array.isArray(gate.coveredPlatforms) || !Array.isArray(gate.coveredProviders)) {
+    return
+  }
+  for (const platform of gate.coveredPlatforms) {
+    if (!asArray(gate.platforms).includes(platform)) {
+      failures.push(`${gate.id}: covered platform is outside risk scope: ${platform}`)
+    }
+  }
+  for (const provider of gate.coveredProviders) {
+    if (!asArray(gate.providers).includes(provider)) {
+      failures.push(`${gate.id}: covered provider is outside risk scope: ${provider}`)
+    }
+  }
+  if (
+    ['partial', 'active'].includes(gate.protection) &&
+    gate.evidenceRuns.some((run) => isRecord(run) && run.result === 'passed')
+  ) {
+    const evidencePlatforms = new Set(
+      gate.evidenceRuns
+        .filter((run) => isRecord(run) && run.result === 'passed')
+        .map((run) => run.platform)
+    )
+    for (const platform of evidencePlatforms) {
+      if (!gate.coveredPlatforms.includes(platform)) {
+        failures.push(
+          `${gate.id}: coveredPlatforms must include passed evidence platform ${platform}`
+        )
+      }
+    }
+  }
+  if (
+    gate.protection === 'none' &&
+    (gate.coveredPlatforms.length > 0 || gate.coveredProviders.length > 0)
+  ) {
+    failures.push(`${gate.id}: protection none gates must not declare covered scope`)
+  }
+}
+
+async function fileExists(root, filePath) {
+  try {
+    const stat = await fs.stat(path.join(root, filePath))
+    return stat.isFile()
+  } catch {
+    return false
+  }
+}
+
+async function validateGate(gate, maturities, root) {
+  const failures = []
+  if (!isRecord(gate)) {
+    return ['gate entry must be an object']
+  }
+  for (const field of ['id', 'title', 'owner', 'layer', 'invariant', 'oracle', 'demotionRule']) {
+    requireNonEmptyString(gate, field, failures)
+  }
+  if (!maturities.has(gate.maturity)) {
+    failures.push(`${gate.id ?? '<unknown>'}: maturity is invalid`)
+  }
+  for (const field of [
+    'surfaces',
+    'platforms',
+    'providers',
+    'motivatingLinks',
+    'promotionCriteria'
+  ]) {
+    requireStringArray(gate, field, failures)
+  }
+  if (!Array.isArray(gate.commands) || !gate.commands.every(isNonEmptyString)) {
+    failures.push(`${gate.id}: commands must be an array of strings`)
+  } else if (gate.commands.some(commandUsesBrittleTestSelector)) {
+    failures.push(
+      `${gate.id}: commands must not rely on title selectors (-t, --grep, or --testNamePattern)`
+    )
+  }
+  if (!Array.isArray(gate.testFiles) || !gate.testFiles.every(isNonEmptyString)) {
+    failures.push(`${gate.id}: testFiles must be an array of strings`)
+  } else {
+    for (const testFile of gate.testFiles) {
+      if (!(await fileExists(root, testFile))) {
+        failures.push(`${gate.id}: test file does not exist: ${testFile}`)
+      }
+    }
+  }
+  validateCommandCoverage(gate, failures)
+  if (['soak', 'blocking'].includes(gate.maturity)) {
+    if (!hasNonEmptyStringArray(gate.commands)) {
+      failures.push(`${gate.id}: ${gate.maturity} gates must declare at least one command`)
+    }
+    if (!hasNonEmptyStringArray(gate.testFiles)) {
+      failures.push(`${gate.id}: ${gate.maturity} gates must declare at least one test file`)
+    }
+    if (!['soaking', 'stable'].includes(gate.flakeHistory?.status)) {
+      failures.push(`${gate.id}: ${gate.maturity} gates must have soaking or stable flakeHistory`)
+    }
+    if (!hasCompleteRedGreenEvidence(gate)) {
+      failures.push(`${gate.id}: ${gate.maturity} gates must have complete red/green evidence`)
+    }
+  }
+  validateRuntimeBudget(gate, failures)
+  validateEvidence(gate, 'flakeHistory', FLAKE_STATUSES, failures)
+  validateEvidence(gate, 'redGreenEvidence', RED_GREEN_STATUSES, failures)
+  validateProtection(gate, failures)
+  validateEvidenceRuns(gate, failures)
+  validateAssertionRefs(gate, failures)
+  validateCoveredScope(gate, failures)
+  validatePerformanceBudget(gate, failures)
+  if (!Array.isArray(gate.knownGaps) || !gate.knownGaps.every(isNonEmptyString)) {
+    failures.push(`${gate.id}: knownGaps must be an array of strings`)
+  }
+  return failures
+}
+
+export async function main(root = process.cwd()) {
+  const manifestPath = path.join(root, MANIFEST_PATH)
+  let raw
+  try {
+    raw = await fs.readFile(manifestPath, 'utf8')
+  } catch (error) {
+    const message = error instanceof Error ? error.message : String(error)
+    console.error(`${MANIFEST_PATH}: unable to read manifest (${message})`)
+    return 1
+  }
+  const parseErrors = []
+  const manifest = parse(raw, parseErrors, { allowTrailingComma: true })
+  if (parseErrors.length > 0) {
+    for (const error of parseErrors) {
+      console.error(
+        `${MANIFEST_PATH}: JSONC parse error ${printParseErrorCode(error.error)} at offset ${error.offset}`
+      )
+    }
+    return 1
+  }
+  const failures = []
+  if (!isRecord(manifest)) {
+    failures.push('manifest must be an object')
+  } else {
+    if (manifest.schemaVersion !== 1) {
+      failures.push('schemaVersion must be 1')
+    }
+    const maturities = validatePolicy(manifest, failures)
+    if (!Array.isArray(manifest.gates) || manifest.gates.length === 0) {
+      failures.push('gates must be a non-empty array')
+    } else {
+      const seenIds = new Set()
+      for (const gate of manifest.gates) {
+        if (isRecord(gate) && isNonEmptyString(gate.id)) {
+          if (seenIds.has(gate.id)) {
+            failures.push(`${gate.id}: duplicate gate id`)
+          }
+          seenIds.add(gate.id)
+        }
+        failures.push(...(await validateGate(gate, maturities, root)))
+      }
+    }
+  }
+  if (failures.length > 0) {
+    console.error(`Reliability gate manifest check failed with ${failures.length} issue(s):`)
+    for (const failure of failures) {
+      console.error(`- ${failure}`)
+    }
+    return 1
+  }
+  console.log(`Reliability gate manifest check passed for ${manifest.gates.length} gate(s).`)
+  return 0
+}
+
+if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) {
+  process.exit(await main())
+}

config/scripts/check-reliability-gates.test.mjs

@@ -0,0 +1,460 @@
+import { mkdirSync, mkdtempSync, rmSync, writeFileSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import path from 'node:path'
+import { afterEach, describe, expect, it, vi } from 'vitest'
+import { main } from './check-reliability-gates.mjs'
+
+const tempDirs = []
+
+function makeTempRoot(manifest) {
+  const root = mkdtempSync(path.join(tmpdir(), 'orca-reliability-gates-'))
+  tempDirs.push(root)
+  const configDir = path.join(root, 'config')
+  mkdirSync(configDir, { recursive: true })
+  writeFileSync(path.join(configDir, 'reliability-gates.jsonc'), manifest, 'utf8')
+  writeFileSync(path.join(root, 'some.test.ts'), '', 'utf8')
+  return root
+}
+
+function validManifest(overrides = {}) {
+  return JSON.stringify(
+    {
+      schemaVersion: 1,
+      policy: {
+        maturityLevels: ['experimental', 'soak', 'blocking', 'accepted-gap', 'deprecated'],
+        blockingPromotion: {
+          minimumSoakRuns: 100,
+          minimumSoakDays: 14,
+          maximumUnexplainedFlakes: 0
+        }
+      },
+      gates: [
+        {
+          id: 'terminal-session.snapshot-freshness',
+          title: 'Stale liveness snapshots cannot close newer PTY bindings',
+          maturity: 'soak',
+          protection: 'partial',
+          owner: 'terminal-runtime',
+          layer: 'renderer-unit',
+          surfaces: ['terminal lifecycle'],
+          platforms: ['macos', 'linux', 'windows'],
+          providers: ['local', 'daemon'],
+          coveredPlatforms: ['macos'],
+          coveredProviders: ['local'],
+          coverageNotes: 'One local macOS unit run covers the local decision-layer fixture only.',
+          motivatingLinks: ['https://github.com/stablyai/orca/issues/6773'],
+          invariant: 'A stale snapshot cannot close a newer binding.',
+          oracle: 'The test rejects reconciliation when the binding is newer than the snapshot.',
+          commands: ['pnpm exec vitest run some.test.ts'],
+          testFiles: ['some.test.ts'],
+          assertionRefs: [
+            {
+              file: 'some.test.ts',
+              assertions: ['stale liveness snapshots do not close newer PTY bindings']
+            }
+          ],
+          evidenceRuns: [
+            {
+              date: '2026-07-02',
+              runner: 'local',
+              platform: 'macos',
+              command: 'pnpm exec vitest run some.test.ts',
+              result: 'passed',
+              durationSeconds: 1.2,
+              summary: '1 file passed.'
+            }
+          ],
+          runtimeBudget: {
+            p95Seconds: 10,
+            scope: 'local unit test'
+          },
+          flakeHistory: {
+            status: 'soaking',
+            evidence: 'Soak history exists.'
+          },
+          redGreenEvidence: {
+            status: 'complete',
+            evidence: 'Fails when the guard is removed and passes with the fix.'
+          },
+          performanceBudget: {
+            required: true,
+            evidence: 'Perf measurement is required before blocking promotion.'
+          },
+          promotionCriteria: ['Collect soak history.'],
+          knownGaps: ['Needs an Electron survival test.'],
+          demotionRule: 'Demote on unexplained flakes.',
+          ...overrides
+        }
+      ]
+    },
+    null,
+    2
+  )
+}
+
+afterEach(() => {
+  vi.restoreAllMocks()
+  while (tempDirs.length > 0) {
+    rmSync(tempDirs.pop(), { force: true, recursive: true })
+  }
+})
+
+describe('check-reliability-gates', () => {
+  it('accepts the checked-in manifest', async () => {
+    vi.spyOn(console, 'log').mockImplementation(() => {})
+
+    await expect(main(process.cwd())).resolves.toBe(0)
+  })
+
+  it('rejects soak gates without executable commands', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ commands: [] }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('Reliability gate manifest check failed')
+    )
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('soak gates must declare at least one command')
+    )
+  })
+
+  it('rejects soak gates without test files', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ testFiles: [] }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('soak gates must declare at least one test file')
+    )
+  })
+
+  it('rejects command-backed gates that select tests by title', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({ maturity: 'experimental', commands: ['pnpm vitest -t flaky-title'] })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('commands must not rely on title selectors')
+    )
+  })
+
+  it('rejects gates without a protection level', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const manifest = JSON.parse(validManifest())
+    delete manifest.gates[0].protection
+    const root = makeTempRoot(JSON.stringify(manifest, null, 2))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection must be one of none, partial, or active')
+    )
+  })
+
+  it('rejects commandless gates unless they declare protection none', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        protection: 'partial',
+        commands: [],
+        testFiles: [],
+        flakeHistory: { status: 'not-started', evidence: 'Planning entry only.' },
+        redGreenEvidence: { status: 'missing', evidence: 'No executable proof yet.' }
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('commandless gates must declare protection none')
+    )
+  })
+
+  it('rejects command-backed gates marked as no protection', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ maturity: 'experimental', protection: 'none' }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection none gates must not declare commands or testFiles')
+    )
+  })
+
+  it('reserves active protection for stable blocking gates', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ maturity: 'experimental', protection: 'active' }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection active is reserved for blocking gates')
+    )
+  })
+
+  it('rejects partial gates without a passed evidence run', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ maturity: 'experimental', evidenceRuns: [] }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection partial gates need a passed evidence run')
+    )
+  })
+
+  it('rejects evidence runs whose command is not declared by the gate', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        evidenceRuns: [
+          {
+            date: '2026-07-02',
+            runner: 'local',
+            platform: 'macos',
+            command: 'pnpm exec vitest run other.test.ts',
+            result: 'passed',
+            durationSeconds: 1,
+            summary: 'Wrong command.'
+          }
+        ]
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('evidenceRuns[0].command must match one of the gate commands')
+    )
+  })
+
+  it('rejects protection none gates with evidence runs', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        protection: 'none',
+        commands: [],
+        testFiles: [],
+        flakeHistory: { status: 'not-started', evidence: 'Planning entry only.' },
+        redGreenEvidence: { status: 'missing', evidence: 'No executable proof yet.' }
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection none gates must not declare evidenceRuns')
+    )
+  })
+
+  it('rejects partial gates without assertion refs', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(validManifest({ maturity: 'experimental', assertionRefs: [] }))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection partial gates need assertionRefs')
+    )
+  })
+
+  it('rejects assertion refs outside declared test files', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        assertionRefs: [
+          {
+            file: 'other.test.ts',
+            assertions: ['important invariant']
+          }
+        ]
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('assertionRefs[0].file must be one of the gate testFiles')
+    )
+  })
+
+  it('rejects protection none gates with assertion refs', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        protection: 'none',
+        commands: [],
+        testFiles: [],
+        evidenceRuns: [],
+        assertionRefs: [
+          {
+            file: 'some.test.ts',
+            assertions: ['planning entry must not look tested']
+          }
+        ],
+        coveredPlatforms: [],
+        coveredProviders: [],
+        flakeHistory: { status: 'not-started', evidence: 'Planning entry only.' },
+        redGreenEvidence: { status: 'missing', evidence: 'No executable proof yet.' }
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection none gates must not declare assertionRefs')
+    )
+  })
+
+  it('rejects covered scope outside the risk scope', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        coveredPlatforms: ['ios'],
+        coveredProviders: ['ssh']
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('covered platform is outside risk scope: ios')
+    )
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('covered provider is outside risk scope: ssh')
+    )
+  })
+
+  it('requires coveredPlatforms to include passed evidence platforms', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        coveredPlatforms: ['linux']
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('coveredPlatforms must include passed evidence platform macos')
+    )
+  })
+
+  it('rejects covered scope on protection none gates', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        protection: 'none',
+        commands: [],
+        testFiles: [],
+        evidenceRuns: [],
+        coveredPlatforms: ['macos'],
+        coveredProviders: [],
+        flakeHistory: { status: 'not-started', evidence: 'Planning entry only.' },
+        redGreenEvidence: { status: 'missing', evidence: 'No executable proof yet.' }
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('protection none gates must not declare covered scope')
+    )
+  })
+
+  it('rejects declared test files that do not exist', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({ maturity: 'experimental', testFiles: ['missing.test.ts'] })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('test file does not exist: missing.test.ts')
+    )
+  })
+
+  it('rejects executable gates whose commands do not reference declared test files', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const command = 'pnpm exec vitest run'
+    const root = makeTempRoot(
+      validManifest({
+        maturity: 'experimental',
+        commands: [command],
+        evidenceRuns: [
+          {
+            date: '2026-07-02',
+            runner: 'local',
+            platform: 'macos',
+            command,
+            result: 'passed',
+            durationSeconds: 1,
+            summary: 'Command did not name the declared file.'
+          }
+        ]
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('test file is not referenced by any gate command: some.test.ts')
+    )
+  })
+
+  it('rejects soak gates without mature flake and red/green evidence', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot(
+      validManifest({
+        flakeHistory: { status: 'unknown', evidence: 'No soak yet.' },
+        redGreenEvidence: { status: 'partial', evidence: 'No saved artifact yet.' }
+      })
+    )
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('soak gates must have soaking or stable flakeHistory')
+    )
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining('soak gates must have complete red/green evidence')
+    )
+  })
+
+  it('rejects malformed JSONC', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = makeTempRoot('{ "schemaVersion": 1, } trailing')
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(expect.stringContaining('JSONC parse error'))
+  })
+
+  it('rejects missing manifests with a structured error', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const root = mkdtempSync(path.join(tmpdir(), 'orca-reliability-gates-'))
+    tempDirs.push(root)
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(expect.stringContaining('unable to read manifest'))
+  })
+
+  it('uses policy maturity levels as the source of truth', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const manifest = JSON.parse(validManifest())
+    manifest.policy.maturityLevels = ['experimental']
+    const root = makeTempRoot(JSON.stringify(manifest, null, 2))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(expect.stringContaining('maturity is invalid'))
+  })
+
+  it('rejects malformed blocking promotion policy', async () => {
+    vi.spyOn(console, 'error').mockImplementation(() => {})
+    const manifest = JSON.parse(validManifest())
+    manifest.policy.blockingPromotion.maximumUnexplainedFlakes = -1
+    const root = makeTempRoot(JSON.stringify(manifest, null, 2))
+
+    await expect(main(root)).resolves.toBe(1)
+    expect(console.error).toHaveBeenCalledWith(
+      expect.stringContaining(
+        'policy.blockingPromotion.maximumUnexplainedFlakes must be a non-negative number'
+      )
+    )
+  })
+})

config/scripts/computer-e2e-workflow.test.mjs

@@ -110,6 +110,39 @@ describe('computer-use e2e workflow', () => {
     }
   })
 
+  it('boots the built daemon under plain Node in the PR native-smoke job after the main build', () => {
+    const workflow = parse(
+      readFileSync(join(projectDir, '.github/workflows/computer-e2e.yml'), 'utf8')
+    )
+    const steps = workflow.jobs['native-smoke'].steps
+    const runs = steps.map((step) => step.run).filter((run) => typeof run === 'string')
+    const buildIndex = runs.indexOf('pnpm build:electron-vite')
+    const daemonSmokeIndex = runs.indexOf('node config/scripts/daemon-boot-smoke.mjs')
+
+    expect(daemonSmokeIndex, 'native-smoke must boot the built daemon').toBeGreaterThanOrEqual(0)
+    expect(
+      buildIndex,
+      'daemon boot smoke must run after the main bundle is built'
+    ).toBeGreaterThanOrEqual(0)
+    expect(daemonSmokeIndex).toBeGreaterThan(buildIndex)
+  })
+
+  it('re-runs the native-smoke job when the daemon bundle graph changes', () => {
+    const workflow = parse(
+      readFileSync(join(projectDir, '.github/workflows/computer-e2e.yml'), 'utf8')
+    )
+    const triggerPaths = workflow.on.pull_request.paths
+
+    expect(triggerPaths).toEqual(
+      expect.arrayContaining([
+        'config/scripts/daemon-boot-smoke.mjs',
+        'electron.vite.config.ts',
+        'build-plugins/**',
+        'src/main/daemon/**'
+      ])
+    )
+  })
+
   it('runs Linux computer-use e2e in the PR native-smoke job under Xvfb', () => {
     const workflow = parse(
       readFileSync(join(projectDir, '.github/workflows/computer-e2e.yml'), 'utf8')

config/scripts/create-draft-release.mjs

@@ -39,17 +39,21 @@ function compareDesktopReleaseTags(a, b) {
   return a.rc - b.rc
 }
 
-export function latestPreviousDesktopReleaseTag(tags, tag) {
+export function latestPreviousPublishedDesktopReleaseTag(releases, tag) {
   const current = parseDesktopReleaseTag(tag)
   if (!current) {
     return ''
   }
-  const previousTags = tags
-    .map((candidate) => parseDesktopReleaseTag(candidate))
+  const previousReleases = releases
+    .filter((release) => release?.draft === false && typeof release.tag_name === 'string')
+    .map((release) => parseDesktopReleaseTag(release.tag_name))
     .filter((candidate) => candidate && candidate.tag !== current.tag)
     .filter((candidate) => compareDesktopReleaseTags(candidate, current) < 0)
+    // Why: public changelogs should be bounded by releases users could see;
+    // stable releases summarize since the prior stable, not the latest RC.
+    .filter((candidate) => current.rc !== null || candidate.rc === null)
     .sort(compareDesktopReleaseTags)
-  return previousTags.at(-1)?.tag ?? ''
+  return previousReleases.at(-1)?.tag ?? ''
 }
 
 function githubHeaders(token) {
@@ -75,27 +79,23 @@ async function githubJson(fetchImpl, url, token, options = {}) {
   return res.json()
 }
 
-async function fetchRepoTags(repo, token, fetchImpl) {
-  const tags = []
+async function fetchRepoReleases(repo, token, fetchImpl) {
+  const releases = []
   for (let page = 1; ; page += 1) {
-    const pageTags = await githubJson(
+    const pageReleases = await githubJson(
       fetchImpl,
-      `https://api.github.com/repos/${repo}/tags?per_page=100&page=${page}`,
+      `https://api.github.com/repos/${repo}/releases?per_page=100&page=${page}`,
       token
     )
-    if (!Array.isArray(pageTags)) {
-      throw new Error(`GitHub tags response page ${page} for ${repo} was not an array`)
+    if (!Array.isArray(pageReleases)) {
+      throw new Error(`GitHub releases response page ${page} for ${repo} was not an array`)
     }
-    for (const tag of pageTags) {
-      if (typeof tag?.name === 'string') {
-        tags.push(tag.name)
-      }
-    }
-    if (pageTags.length < 100) {
+    releases.push(...pageReleases)
+    if (pageReleases.length < 100) {
       break
     }
   }
-  return tags
+  return releases
 }
 
 export function truncateReleaseBody(body, maxLength = MAX_RELEASE_BODY_LENGTH) {
@@ -128,8 +128,8 @@ export async function createDraftRelease({
     throw new Error('token is required')
   }
 
-  const previousTag = latestPreviousDesktopReleaseTag(
-    await fetchRepoTags(repo, token, fetchImpl),
+  const previousTag = latestPreviousPublishedDesktopReleaseTag(
+    await fetchRepoReleases(repo, token, fetchImpl),
     tag
   )
   const generateNotesBody = {
@@ -138,9 +138,8 @@ export async function createDraftRelease({
     ...(previousTag ? { previous_tag_name: previousTag } : {})
   }
 
-  // Why: draft releases are invisible to GitHub's generate-notes baseline.
-  // Passing the previous desktop tag keeps each release from accumulating
-  // notes from older drafts.
+  // Why: GitHub's generate-notes baseline ignores draft releases, so pass the
+  // previous public changelog boundary explicitly.
   const releaseNotes = await githubJson(
     fetchImpl,
     `https://api.github.com/repos/${repo}/releases/generate-notes`,

config/scripts/create-draft-release.test.mjs

@@ -1,11 +1,19 @@
 import { describe, expect, it, vi } from 'vitest'
 import {
   createDraftRelease,
-  latestPreviousDesktopReleaseTag,
+  latestPreviousPublishedDesktopReleaseTag,
   parseDesktopReleaseTag,
   truncateReleaseBody
 } from './create-draft-release.mjs'
 
+function release(tag, options = {}) {
+  return {
+    draft: false,
+    tag_name: tag,
+    ...options
+  }
+}
+
 function jsonResponse(body, init = {}) {
   return {
     ok: init.ok ?? true,
@@ -49,32 +57,74 @@ describe('parseDesktopReleaseTag', () => {
   })
 })
 
-describe('latestPreviousDesktopReleaseTag', () => {
-  it('bounds stable notes to the previous rc when one exists', () => {
-    expect(latestPreviousDesktopReleaseTag(['v1.4.35', 'v1.4.36-rc.0', 'v1.4.36'], 'v1.4.36')).toBe(
-      'v1.4.36-rc.0'
-    )
+describe('latestPreviousPublishedDesktopReleaseTag', () => {
+  it('bounds stable notes to the previous stable release when rcs exist', () => {
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.35'), release('v1.4.36-rc.0'), release('v1.4.36')],
+        'v1.4.36'
+      )
+    ).toBe('v1.4.35')
+  })
+
+  it('does not collapse a stable changelog to its rc-to-stable version bump', () => {
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [
+          release('v1.4.120'),
+          release('v1.4.121-rc.0'),
+          release('v1.4.121-rc.6'),
+          release('v1.4.121')
+        ],
+        'v1.4.121'
+      )
+    ).toBe('v1.4.120')
   })
 
   it('bounds the first rc notes to the previous stable release', () => {
     expect(
-      latestPreviousDesktopReleaseTag(['v1.4.35', 'v1.4.36-rc.0', 'mobile-v0.0.12'], 'v1.4.36-rc.0')
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.35'), release('v1.4.36-rc.0'), release('mobile-v0.0.12')],
+        'v1.4.36-rc.0'
+      )
     ).toBe('v1.4.35')
   })
 
   it('bounds later rc notes to the prior rc', () => {
-    expect(latestPreviousDesktopReleaseTag(['v1.4.36-rc.0', 'v1.4.36-rc.1'], 'v1.4.36-rc.1')).toBe(
-      'v1.4.36-rc.0'
-    )
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.36-rc.0'), release('v1.4.36-rc.1')],
+        'v1.4.36-rc.1'
+      )
+    ).toBe('v1.4.36-rc.0')
+  })
+
+  it('ignores draft releases as public changelog boundaries', () => {
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.35'), release('v1.4.36-rc.0', { draft: true }), release('v1.4.36-rc.1')],
+        'v1.4.36-rc.1'
+      )
+    ).toBe('v1.4.35')
   })
 
   it('returns empty string for the first desktop release when no earlier tag exists', () => {
-    expect(latestPreviousDesktopReleaseTag(['v1.4.36', 'mobile-v0.0.12'], 'v1.4.36')).toBe('')
-    expect(latestPreviousDesktopReleaseTag([], 'v1.4.36')).toBe('')
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.36'), release('mobile-v0.0.12')],
+        'v1.4.36'
+      )
+    ).toBe('')
+    expect(latestPreviousPublishedDesktopReleaseTag([], 'v1.4.36')).toBe('')
   })
 
   it('returns empty string when the current tag is not a desktop release tag', () => {
-    expect(latestPreviousDesktopReleaseTag(['v1.4.35', 'v1.4.36'], 'mobile-v0.0.12')).toBe('')
+    expect(
+      latestPreviousPublishedDesktopReleaseTag(
+        [release('v1.4.35'), release('v1.4.36')],
+        'mobile-v0.0.12'
+      )
+    ).toBe('')
   })
 })
 
@@ -82,7 +132,7 @@ describe('createDraftRelease', () => {
   it('creates a draft release with bounded generated notes', async () => {
     const fetchImpl = vi
       .fn()
-      .mockResolvedValueOnce(jsonResponse([{ name: 'v1.4.35' }, { name: 'v1.4.36' }]))
+      .mockResolvedValueOnce(jsonResponse([release('v1.4.35'), release('v1.4.36')]))
       .mockResolvedValueOnce(jsonResponse({ name: 'v1.4.36', body: 'a'.repeat(130_000) }))
       .mockResolvedValueOnce(jsonResponse({ tag_name: 'v1.4.36', draft: true }))
 
@@ -96,7 +146,7 @@ describe('createDraftRelease', () => {
 
     expect(fetchImpl).toHaveBeenNthCalledWith(
       1,
-      'https://api.github.com/repos/stablyai/orca/tags?per_page=100&page=1',
+      'https://api.github.com/repos/stablyai/orca/releases?per_page=100&page=1',
       expect.any(Object)
     )
     expect(fetchImpl).toHaveBeenNthCalledWith(
@@ -134,7 +184,7 @@ describe('createDraftRelease', () => {
   it('marks rc tags as prereleases', async () => {
     const fetchImpl = vi
       .fn()
-      .mockResolvedValueOnce(jsonResponse([{ name: 'v1.4.36' }, { name: 'v1.4.36-rc.1' }]))
+      .mockResolvedValueOnce(jsonResponse([release('v1.4.36'), release('v1.4.36-rc.1')]))
       .mockResolvedValueOnce(jsonResponse({ name: 'v1.4.36-rc.1', body: 'notes' }))
       .mockResolvedValueOnce(jsonResponse({ tag_name: 'v1.4.36-rc.1', draft: true }))
 
@@ -153,7 +203,7 @@ describe('createDraftRelease', () => {
   it('omits previous_tag_name for the first desktop release so notes fall back to the GitHub default', async () => {
     const fetchImpl = vi
       .fn()
-      .mockResolvedValueOnce(jsonResponse([{ name: 'v1.4.36' }, { name: 'mobile-v0.0.12' }]))
+      .mockResolvedValueOnce(jsonResponse([release('v1.4.36'), release('mobile-v0.0.12')]))
       .mockResolvedValueOnce(jsonResponse({ name: 'v1.4.36', body: 'notes' }))
       .mockResolvedValueOnce(jsonResponse({ tag_name: 'v1.4.36', draft: true }))
 
@@ -170,12 +220,12 @@ describe('createDraftRelease', () => {
     expect(generateNotesBody).not.toHaveProperty('previous_tag_name')
   })
 
-  it('paginates through every tag page before choosing the previous tag', async () => {
-    const firstPage = Array.from({ length: 100 }, (_, index) => ({ name: `mobile-v0.0.${index}` }))
+  it('paginates through every release page before choosing the previous release', async () => {
+    const firstPage = Array.from({ length: 100 }, (_, index) => release(`mobile-v0.0.${index}`))
     const fetchImpl = vi
       .fn()
       .mockResolvedValueOnce(jsonResponse(firstPage))
-      .mockResolvedValueOnce(jsonResponse([{ name: 'v1.4.35' }]))
+      .mockResolvedValueOnce(jsonResponse([release('v1.4.35')]))
       .mockResolvedValueOnce(jsonResponse({ name: 'v1.4.36', body: 'notes' }))
       .mockResolvedValueOnce(jsonResponse({ tag_name: 'v1.4.36', draft: true }))
 
@@ -189,12 +239,12 @@ describe('createDraftRelease', () => {
 
     expect(fetchImpl).toHaveBeenNthCalledWith(
       1,
-      'https://api.github.com/repos/stablyai/orca/tags?per_page=100&page=1',
+      'https://api.github.com/repos/stablyai/orca/releases?per_page=100&page=1',
       expect.any(Object)
     )
     expect(fetchImpl).toHaveBeenNthCalledWith(
       2,
-      'https://api.github.com/repos/stablyai/orca/tags?per_page=100&page=2',
+      'https://api.github.com/repos/stablyai/orca/releases?per_page=100&page=2',
       expect.any(Object)
     )
     const generateNotesBody = JSON.parse(fetchImpl.mock.calls[2][1].body)

config/scripts/daemon-boot-smoke.mjs

@@ -0,0 +1,217 @@
+/**
+ * Boots the BUILT terminal daemon (out/main/daemon-entry.js) under plain Node —
+ * the exact way production forks it (ELECTRON_RUN_AS_NODE = a plain-Node
+ * process) — and asserts it starts, serves a real PTY, and stops.
+ *
+ * Why this exists: native-smoke CI (and packaging) went green while
+ * v1.4.129-rc.1 shipped a daemon that exited code 1 at module load because an
+ * electron `require` leaked into its bundle graph. Nothing executed the built
+ * entry under plain Node, so the outage was invisible until an adopted old
+ * daemon died in the field. This runs on every PR that touches the daemon.
+ *
+ * Hard assertions (fail the job):
+ *   - the daemon signals `{ type: 'ready' }` over IPC within the timeout, and
+ *   - it terminates when asked (no hang / zombie).
+ * Best-effort (logged skip, never fails): an end-to-end `ptySpawnHealth` RPC,
+ * because node-pty spawn can be flaky on constrained CI runners.
+ */
+import { fork } from 'node:child_process'
+import { connect } from 'node:net'
+import { randomUUID } from 'node:crypto'
+import { mkdtempSync, readFileSync, rmSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import { join, resolve } from 'node:path'
+
+const projectDir = resolve(import.meta.dirname, '../..')
+const entryPath = join(projectDir, 'out', 'main', 'daemon-entry.js')
+
+const READY_TIMEOUT_MS = 30_000
+const PTY_HEALTH_TIMEOUT_MS = 10_000
+const SHUTDOWN_TIMEOUT_MS = 10_000
+
+function log(message) {
+  process.stdout.write(`[daemon-boot-smoke] ${message}\n`)
+}
+
+// Why: the daemon rejects a hello whose protocol version differs, so read the
+// current version from source rather than hardcoding a number that can drift.
+function readProtocolVersion() {
+  const source = readFileSync(join(projectDir, 'src/main/daemon/types.ts'), 'utf8')
+  const match = source.match(/PROTOCOL_VERSION\s*=\s*(\d+)/)
+  if (!match) {
+    throw new Error('could not read PROTOCOL_VERSION from src/main/daemon/types.ts')
+  }
+  return Number(match[1])
+}
+
+function makeSocketPath(userDataDir) {
+  // Why: Windows AF_UNIX-style IPC uses named pipes; POSIX uses a filesystem
+  // socket kept under the scratch userData dir so cleanup removes it.
+  if (process.platform === 'win32') {
+    return `\\\\.\\pipe\\orca-daemon-smoke-${process.pid}-${randomUUID()}`
+  }
+  return join(userDataDir, 'daemon.sock')
+}
+
+// Best-effort end-to-end PTY check over the daemon's own control-socket RPC.
+// Connects a single control socket, completes the hello handshake, and calls
+// `ptySpawnHealth` (the daemon spawns a throwaway PTY internally). Resolves
+// true on success, false on any failure — never throws.
+function runPtySpawnHealthCheck(socketPath, tokenPath, protocolVersion) {
+  return new Promise((resolveCheck) => {
+    let settled = false
+    let buffer = ''
+    const socket = connect(socketPath)
+    const finish = (ok, reason) => {
+      if (settled) {
+        return
+      }
+      settled = true
+      clearTimeout(timer)
+      socket.destroy()
+      if (!ok && reason) {
+        log(`PTY spawn health check skipped (best-effort): ${reason}`)
+      }
+      resolveCheck(ok)
+    }
+    const timer = setTimeout(() => finish(false, 'timed out'), PTY_HEALTH_TIMEOUT_MS)
+
+    socket.on('error', (err) => finish(false, err.message))
+    socket.on('connect', () => {
+      const token = readFileSync(tokenPath, 'utf8').trim()
+      socket.write(
+        `${JSON.stringify({
+          type: 'hello',
+          version: protocolVersion,
+          token,
+          clientId: randomUUID(),
+          role: 'control'
+        })}\n`
+      )
+    })
+    socket.on('data', (chunk) => {
+      buffer += chunk.toString('utf8')
+      let newlineIdx = buffer.indexOf('\n')
+      while (newlineIdx !== -1) {
+        const line = buffer.slice(0, newlineIdx)
+        buffer = buffer.slice(newlineIdx + 1)
+        let msg
+        try {
+          msg = JSON.parse(line)
+        } catch {
+          finish(false, 'invalid response line')
+          return
+        }
+        if (msg.type === 'hello') {
+          if (!msg.ok) {
+            finish(false, `hello rejected: ${msg.error ?? 'unknown'}`)
+            return
+          }
+          socket.write(`${JSON.stringify({ id: 'health-1', type: 'ptySpawnHealth' })}\n`)
+        } else if (msg.id === 'health-1') {
+          finish(
+            msg.ok === true,
+            msg.ok === true ? undefined : (msg.error ?? 'ptySpawnHealth failed')
+          )
+          return
+        }
+        newlineIdx = buffer.indexOf('\n')
+      }
+    })
+  })
+}
+
+async function main() {
+  const userDataDir = mkdtempSync(join(tmpdir(), 'orca-daemon-boot-smoke-'))
+  const socketPath = makeSocketPath(userDataDir)
+  const tokenPath = join(userDataDir, 'daemon.token')
+  const protocolVersion = readProtocolVersion()
+
+  log(`forking ${entryPath} under plain Node (${process.execPath})`)
+  const child = fork(entryPath, ['--socket', socketPath, '--token', tokenPath], {
+    // Plain Node: no ELECTRON_RUN_AS_NODE. process.execPath is already node in
+    // CI, and this is exactly the runtime where a leaked `require("electron")`
+    // throws MODULE_NOT_FOUND — the failure this smoke exists to catch.
+    stdio: ['ignore', 'pipe', 'pipe', 'ipc'],
+    env: { ...process.env, ORCA_USER_DATA_PATH: userDataDir }
+  })
+
+  let stderr = ''
+  child.stderr?.on('data', (chunk) => {
+    stderr += chunk.toString('utf8')
+  })
+  child.stdout?.on('data', (chunk) => {
+    process.stdout.write(chunk)
+  })
+
+  const cleanup = () => {
+    if (child.exitCode === null && child.signalCode === null && child.pid) {
+      try {
+        child.kill('SIGKILL')
+      } catch {
+        // already gone
+      }
+    }
+    rmSync(userDataDir, { recursive: true, force: true })
+  }
+
+  try {
+    await new Promise((resolveReady, rejectReady) => {
+      const timer = setTimeout(() => {
+        rejectReady(
+          new Error(
+            `daemon did not signal 'ready' within ${READY_TIMEOUT_MS}ms.\nstderr:\n${stderr}`
+          )
+        )
+      }, READY_TIMEOUT_MS)
+      child.on('message', (msg) => {
+        if (msg && typeof msg === 'object' && msg.type === 'ready') {
+          clearTimeout(timer)
+          resolveReady()
+        }
+      })
+      child.on('error', (err) => {
+        clearTimeout(timer)
+        rejectReady(new Error(`daemon fork errored: ${err.message}\nstderr:\n${stderr}`))
+      })
+      child.on('exit', (code, signal) => {
+        clearTimeout(timer)
+        rejectReady(
+          new Error(
+            `daemon exited before 'ready' (code=${code}, signal=${signal}).\nstderr:\n${stderr}`
+          )
+        )
+      })
+    })
+    log('daemon signaled ready')
+
+    const ptyHealthy = await runPtySpawnHealthCheck(socketPath, tokenPath, protocolVersion)
+    if (ptyHealthy) {
+      log('ptySpawnHealth OK — daemon spawned a real PTY end-to-end')
+    }
+
+    await new Promise((resolveExit, rejectExit) => {
+      const timer = setTimeout(() => {
+        rejectExit(new Error(`daemon did not exit within ${SHUTDOWN_TIMEOUT_MS}ms of SIGTERM`))
+      }, SHUTDOWN_TIMEOUT_MS)
+      child.on('exit', (code, signal) => {
+        clearTimeout(timer)
+        log(`daemon exited after signal (code=${code}, signal=${signal})`)
+        resolveExit()
+      })
+      // Why: SIGTERM is the graceful stop on POSIX (the daemon handles it);
+      // Windows has no POSIX signal delivery, so Node maps this to process
+      // termination. Either way the hard assertion is "it stops, no hang".
+      child.kill('SIGTERM')
+    })
+
+    log('PASS: daemon booted, served, and shut down under plain Node')
+  } finally {
+    cleanup()
+  }
+}
+
+main().catch((error) => {
+  process.stderr.write(`[daemon-boot-smoke] FAIL: ${error.message}\n`)
+  process.exitCode = 1
+})

config/scripts/electron-builder-config.test.mjs

@@ -63,12 +63,36 @@ describe('electron-builder config', () => {
     )
   })
 
+  // Why: on macOS 26 UNUserNotificationCenter aborts for executables launched
+  // from Contents/Resources, so the helper must ship in Contents/MacOS (#7929).
+  it('ships the mac notification-status helper in Contents/MacOS, not Resources', () => {
+    expect(electronBuilderConfig.mac.extraFiles).toEqual(
+      expect.arrayContaining([
+        expect.objectContaining({
+          from: 'native/notification-status-macos/.build/release/orca-notification-status',
+          to: 'MacOS/orca-notification-status'
+        })
+      ])
+    )
+    expect(electronBuilderConfig.mac.extraResources).not.toEqual(
+      expect.arrayContaining([expect.objectContaining({ to: 'orca-notification-status' })])
+    )
+  })
+
   it('unpacks the compiled CommonJS boundary with CLI runtime files', () => {
     expect(electronBuilderConfig.asarUnpack).toEqual(
       expect.arrayContaining(['out/package.json', 'out/cli/**', 'out/shared/**'])
     )
   })
 
+  // Why: without the unpacked entry the watcher client silently falls back to
+  // in-process @parcel/watcher, reintroducing the #7547 main-process crash.
+  it('unpacks the forked parcel-watcher process entry', () => {
+    expect(electronBuilderConfig.asarUnpack).toEqual(
+      expect.arrayContaining(['out/main/parcel-watcher-process-entry.js'])
+    )
+  })
+
   it('uses the multi-size icon source for Linux packages', () => {
     expect(electronBuilderConfig.linux.icon).toBe('resources/build/icon.icns')
   })

config/scripts/locale-ko-key-overrides.json

@@ -1559,12 +1559,6 @@
   "auto.components.onboarding.ThemeStep.94b9dc561d": {
     "ko": "설정 -> Terminal"
   },
-  "auto.components.pet.PetOverlay.4712d196c6": {
-    "ko": "@keyframes pet-{{value0}} { from { background-position: {{value1}}px {{value2}}px; } to { background-position: {{value3}}px {{value4}}px; } }"
-  },
-  "auto.components.pet.PetOverlay.de932b0e8f": {
-    "ko": "@keyframes pet-bob { 0%,100% { transform: translateY(0); } 50% { transform: translateY(-4px); } }"
-  },
   "auto.components.repo.NestedRepoChecklist.ea54c7bf8f": {
     "ko": "/"
   },

config/scripts/orchestration-skill-guidance.test.mjs

@@ -19,6 +19,28 @@ function getSection(markdown, heading) {
 }
 
 describe('orchestration skill guidance', () => {
+  it('requires Orca runtime state before claiming a worker was orchestrated', () => {
+    const skill = readSkill()
+    const toolBoundary = getSection(skill, 'Tool Boundary')
+
+    expect(toolBoundary).toContain(
+      'must create Orca runtime state with `orca orchestration task-create` and `orca orchestration dispatch --inject`'
+    )
+    expect(toolBoundary).toContain('or `orca orchestration run`')
+    expect(toolBoundary).toContain(
+      'Do not substitute non-Orca subagent tools, generic agent-spawn APIs, or chat-only parallel worker features'
+    )
+    expect(toolBoundary).toContain('do not create Orca task/dispatch provenance')
+    expect(toolBoundary).toContain('injected lifecycle preambles')
+    expect(toolBoundary).toContain('`worker_done` authority')
+    expect(toolBoundary).toContain('decision gates')
+    expect(toolBoundary).toContain('orca orchestration task-list --json')
+    expect(toolBoundary).toContain('orca orchestration dispatch-show --task <task_id> --json')
+    expect(toolBoundary).toContain(
+      'do not retroactively describe the external worker as orchestrated'
+    )
+  })
+
   it('treats long-running worker waits as liveness checkpoints, not failures', () => {
     const skill = readSkill()
 
@@ -54,6 +76,17 @@ describe('orchestration skill guidance', () => {
     expect(skill).toContain(
       'orca worktree create --name <task-name> --no-parent --agent codex --prompt'
     )
+    expect(fullHandoffs).toContain(
+      'Before creating a new worktree from an active feature branch, decide and state whether the desired Orca lineage is child or top-level'
+    )
+    expect(fullHandoffs).toContain(
+      'Use child worktree lineage only when the new work is conceptually stacked under or dependent on the active worktree'
+    )
+    expect(fullHandoffs).toContain(
+      'For independent repo-wide fixes, standalone feature work, or unrelated follow-up tasks, create a top-level worktree with `--no-parent`'
+    )
+    expect(fullHandoffs).toContain('If the work should start from the repo default base')
+    expect(fullHandoffs).toContain('omit `--base-branch`')
   })
 
   it('classifies handoff wording as ownership transfer unless supervision is explicit', () => {
@@ -101,6 +134,33 @@ describe('orchestration skill guidance', () => {
     expect(fullHandoffs).toContain('Do not monitor task completion.')
   })
 
+  it('clarifies sidebar lineage for same-worktree orchestrated workers', () => {
+    const skill = readSkill()
+    const workerTerminals = getSection(skill, 'Worker Terminals')
+
+    expect(workerTerminals).toContain(
+      'Sidebar lineage and orchestration lifecycle are related but not identical.'
+    )
+    expect(workerTerminals).toContain(
+      'A same-worktree worker created with `orca terminal create --worktree active` may appear as a peer terminal/agent'
+    )
+    expect(workerTerminals).toContain(
+      'even though it is a child dispatch in Orca orchestration state'
+    )
+    expect(workerTerminals).toContain(
+      'A visible parent/child worktree relationship requires creating a child worktree'
+    )
+    expect(workerTerminals).toContain(
+      'only when the task can safely run from an isolated checkout and does not need uncommitted artifacts from the current working tree'
+    )
+    expect(workerTerminals).toContain(
+      'For supervised new-worktree workers, decide the desired Orca lineage before creation'
+    )
+    expect(workerTerminals).toContain(
+      'use `--no-parent` for independent repo-wide fixes, standalone feature work, or unrelated follow-up tasks'
+    )
+  })
+
   it('keeps review-only completions and named next-owner fixes in their lanes', () => {
     const skill = readSkill()
 
@@ -116,4 +176,16 @@ describe('orchestration skill guidance', () => {
     expect(skill).toContain('post-review corrections and PR prep belong to that named owner')
     expect(skill).toContain('the named owner edits files and creates the PR')
   })
+
+  it('keeps worker_done post-completion guidance idle instead of polling', () => {
+    const skill = readSkill()
+    const agentGuidance = getSection(skill, 'Agent Guidance')
+
+    expect(agentGuidance).toContain('After sending `worker_done`, end your turn')
+    expect(agentGuidance).toContain('idle at the agent prompt')
+    expect(agentGuidance).toContain('Do not poll or keep calling `orca orchestration check`')
+    expect(agentGuidance).toContain('fresh preamble + TASK block delivered as new terminal input')
+    expect(skill).not.toContain('post-completion polling messages')
+    expect(skill).not.toContain('every 2 minutes')
+  })
 })

config/scripts/package-electron-runtime-contract.test.mjs

@@ -221,7 +221,7 @@ describe('Electron runtime package contract', () => {
     expect(installRun).not.toMatch(/throw\s+\$_/)
   })
 
-  it('temporarily allows publishing Windows after verifying the signed installer only', () => {
+  it('verifies Windows inner binary signatures fail-open before publishing', () => {
     const releaseWorkflow = readFileSync(
       join(projectDir, '.github/workflows/release-cut.yml'),
       'utf8'
@@ -230,12 +230,38 @@ describe('Electron runtime package contract', () => {
     const steps = parsedWorkflow.jobs.build.steps
     const stepNames = steps.map((step) => step.name)
     const outerVerifyIndex = stepNames.indexOf('Verify signed Windows installer')
-    const innerVerifyIndex = stepNames.indexOf('Verify signed Windows inner executable')
+    const innerVerifyIndex = stepNames.indexOf('Verify Windows inner binary signatures')
+    const evidenceIndex = stepNames.indexOf('Upload Windows inner signing evidence')
     const publishIndex = stepNames.indexOf('Publish signed Windows release artifacts')
 
     expect(outerVerifyIndex).toBeGreaterThan(-1)
-    expect(innerVerifyIndex).toBe(-1)
-    expect(publishIndex).toBe(outerVerifyIndex + 1)
+    expect(innerVerifyIndex).toBe(outerVerifyIndex + 1)
+    expect(evidenceIndex).toBe(innerVerifyIndex + 1)
+    expect(publishIndex).toBe(evidenceIndex + 1)
+
+    // Why fail-open: unsigned inner binaries must warn, not block, until the
+    // flow is proven on a real release (issue #7785). Flip this to 'true'
+    // together with the workflow env to make the gate required.
+    expect(steps[innerVerifyIndex].env.ORCA_WINDOWS_INNER_SIGNATURE_REQUIRED).toBe('false')
+
+    // Why: every step in the inner-signing chain must be unable to fail the
+    // release — a SignPath outage or timeout falls through to today's
+    // unsigned-inner flow instead of blocking the cut.
+    const innerChainStepNames = [
+      'Stage unsigned inner PE files for signing',
+      'Upload unsigned inner binaries for SignPath',
+      'Submit inner binaries signing request',
+      'Notify Slack that inner-binary signing is waiting for approval',
+      'Download signed inner binaries from SignPath',
+      'Restore signed inner binaries into unpacked app',
+      'Replace cached elevate.exe with the signed copy',
+      'Rebuild NSIS installer from signed unpacked app'
+    ]
+    for (const stepName of innerChainStepNames) {
+      const step = steps[stepNames.indexOf(stepName)]
+      expect(step, stepName).toBeDefined()
+      expect(step['continue-on-error'], stepName).toBe(true)
+    }
   })
 
   it('publishes both Linux release matrix entries', () => {

config/scripts/run-electron-vite-dev.mjs

@@ -50,7 +50,7 @@ function readGitValue(args) {
 }
 
 function lastBranchSegment(value) {
-  return value.replace(/\\/g, '/').split('/').filter(Boolean).at(-1) ?? value
+  return value.replace(/\\/g, '/').split('/').findLast(Boolean) ?? value
 }
 
 function formatDevInstanceLabel(branch, worktreeName) {
@@ -97,16 +97,6 @@ function setPlistValue(plistPath, key, value) {
   execFileSync('/usr/bin/plutil', ['-replace', key, '-string', value, plistPath])
 }
 
-function sanitizeBundleIdPart(value) {
-  return (
-    value
-      .toLowerCase()
-      .replace(/[^a-z0-9.-]+/g, '-')
-      .replace(/^-+|-+$/g, '')
-      .slice(0, 80) || 'dev'
-  )
-}
-
 function sanitizeMacAppBundleName(value) {
   return (
     Array.from(value, (char) => {
@@ -138,7 +128,10 @@ function prepareMacDevElectronApp() {
 
   const title = process.env.ORCA_DEV_DOCK_TITLE || 'Orca: dev'
   const identityKey = process.env.ORCA_DEV_INSTANCE_KEY || repoRoot
-  const bundleLayoutVersion = 'dock-title-app-preserve-framework-symlinks-v4'
+  // v6: bundle the notification-status helper (real permission readout) and
+  // ad-hoc re-sign after plist edits so Notification Center accepts the
+  // bundle; bumping forces stale cached copies to be recreated.
+  const bundleLayoutVersion = 'dock-title-app-preserve-framework-symlinks-v6'
   const hash = createHash('sha1')
     .update(
       `${sourceAppPath}\0${electronVersion ?? ''}\0${title}\0${identityKey}\0${bundleLayoutVersion}`
@@ -151,7 +144,17 @@ function prepareMacDevElectronApp() {
   const appBundleName = `${sanitizeMacAppBundleName(title)}.app`
   const appPath = path.join(distDir, appBundleName)
   const markerPath = path.join(distDir, 'orca-dev-electron-app.json')
-  const bundleId = `com.stablyai.orca.dev.${sanitizeBundleIdPart(hash)}`
+  // Why: one stable id for every dev instance. Per-instance ids registered a
+  // new macOS Notification Settings entry for each branch × Electron version,
+  // piling up "Orca: <branch>" rows forever and breaking the notification
+  // settings deep-link (System Settings can't resolve an id it has no entry
+  // for and falls back to the root list). macOS keys notification permission
+  // by bundle id, so a single id also means granting notifications to one dev
+  // instance covers all of them. Trade-off: when two dev instances run at
+  // once, macOS may route a notification click to the other instance —
+  // Electron drops clicks for notification ids it didn't create, so the
+  // click is lost, not misdirected.
+  const bundleId = 'com.stablyai.orca.dev'
   process.env.ORCA_DEV_MACOS_BUNDLE_ID = bundleId
   const expectedMarker = JSON.stringify(
     { title, appBundleName, bundleId, sourceAppPath, electronVersion, bundleLayoutVersion },
@@ -206,9 +209,45 @@ function prepareMacDevElectronApp() {
   setPlistValue(plistPath, 'CFBundleDisplayName', title)
   setPlistValue(plistPath, 'CFBundleIdentifier', bundleId)
 
-  // Why no re-sign: dev launches execute the copied Electron binary directly,
-  // and Electron's framework bundle is ambiguous to codesign when deep-signing
-  // an already-built distribution. Avoid blocking `pn dev` on local signing.
+  // Why: the notification-status helper reads the app's real macOS
+  // notification authorization (UNUserNotificationCenter has no Electron
+  // API). It must live inside the bundle and carry the dev bundle id as its
+  // embedded/code-sign identifier — macOS keys notification records to the
+  // signing identifier. Non-fatal: without swiftc the permission card falls
+  // back to delivery-probe heuristics.
+  try {
+    execFileSync(
+      process.execPath,
+      [
+        path.join(repoRoot, 'config', 'scripts', 'build-notification-status-macos.mjs'),
+        '--bundle-id',
+        bundleId,
+        '--single-arch',
+        '--output',
+        path.join(appPath, 'Contents', 'MacOS', 'orca-notification-status')
+      ],
+      { stdio: 'inherit' }
+    )
+  } catch (error) {
+    console.warn(
+      `[orca-dev] notification-status helper build failed (permission card falls back to probes): ${error?.message ?? error}`
+    )
+  }
+
+  // Why: the plist edits above (and the copy itself) break the bundle's
+  // ad-hoc seal, and macOS refuses Notification Center registration for
+  // invalidly-signed apps — every dev notification fails with UNErrorDomain
+  // error 1 and the app never appears in System Settings > Notifications.
+  // An ad-hoc re-sign restores delivery, the permission prompt, and the
+  // notification-settings deep link for dev builds. Non-fatal: a signing
+  // failure should not block `pnpm dev`.
+  try {
+    execFileSync('/usr/bin/codesign', ['--force', '--deep', '--sign', '-', appPath])
+  } catch (error) {
+    console.warn(
+      `[orca-dev] ad-hoc codesign failed (dev notifications will not deliver): ${error?.message ?? error}`
+    )
+  }
   writeFileSync(markerPath, expectedMarker, 'utf8')
   process.env.ELECTRON_EXEC_PATH = executablePath
 }
@@ -433,7 +472,7 @@ function isPortFree(port) {
 async function pickDebugPort() {
   // Why: 32 bits of SHA1 (vs 16) reduces truncation bias; modulo 200 still
   // collides routinely across many worktrees, hence the probe sweep below.
-  const seed = parseInt(createHash('sha1').update(repoRoot).digest('hex').slice(0, 8), 16)
+  const seed = Number.parseInt(createHash('sha1').update(repoRoot).digest('hex').slice(0, 8), 16)
   const base = 9333 + (seed % 200) // deterministic base in 9333..9532; probe sweeps up to base+63
   for (let i = 0; i < 64; i++) {
     const p = base + i

config/scripts/serve-headless-fresh-profile-pairing.mjs

@@ -0,0 +1,326 @@
+#!/usr/bin/env node
+
+import { spawn, spawnSync } from 'node:child_process'
+import { existsSync, mkdirSync, mkdtempSync, rmSync } from 'node:fs'
+import { hostname, tmpdir } from 'node:os'
+import path from 'node:path'
+import { createInterface } from 'node:readline'
+
+const scriptDir = import.meta.dirname
+const repoRoot = path.resolve(scriptDir, '..', '..')
+const orcaDevScript = path.join(scriptDir, 'orca-dev.mjs')
+const ensureNativeRuntimeScript = path.join(scriptDir, 'ensure-native-runtime.mjs')
+const fixedProfileDir = process.env.ORCA_HEADLESS_PAIRING_PROFILE_DIR
+const parsed = parseArgs(process.argv.slice(2))
+
+if (parsed.help) {
+  printHelp()
+  process.exit(0)
+}
+
+if (hasForwardedServeFlag(parsed.serveArgs, 'no-pairing')) {
+  console.error('serve-headless-fresh-profile-pairing: --no-pairing cannot print a pairing code.')
+  process.exit(2)
+}
+if (hasForwardedServeFlag(parsed.serveArgs, 'recipe-json')) {
+  console.error(
+    'serve-headless-fresh-profile-pairing: --recipe-json detaches the server and cannot use an ephemeral profile safely.'
+  )
+  process.exit(2)
+}
+
+const serveArgs = withDefaultPairingAddress(parsed.serveArgs)
+
+ensureElectronRuntime()
+
+const profileDir =
+  fixedProfileDir ?? mkdtempSync(path.join(tmpdir(), 'orca-headless-pairing-profile-'))
+const ownsProfileDir = !fixedProfileDir
+mkdirSync(profileDir, { recursive: true })
+
+let cleanedUp = false
+let child = null
+let sawPairingUrl = false
+let stopAttempts = 0
+// Why: temp dev worktrees do not have a root-owned chrome-sandbox; this script
+// is only for local headless testing, not packaged production.
+const childEnv = {
+  ...process.env,
+  ORCA_DEV_USER_DATA_PATH: profileDir,
+  ...(process.platform === 'linux'
+    ? { ELECTRON_DISABLE_SANDBOX: process.env.ELECTRON_DISABLE_SANDBOX ?? '1' }
+    : {})
+}
+
+console.error(`[headless-pairing] userData=${profileDir}`)
+console.error(`[headless-pairing] starting: orca-dev serve --json${formatForwardedArgs(serveArgs)}`)
+
+child = spawn(process.execPath, [orcaDevScript, 'serve', '--json', ...serveArgs], {
+  cwd: repoRoot,
+  detached: process.platform !== 'win32',
+  env: childEnv,
+  stdio: ['inherit', 'pipe', 'inherit']
+})
+
+const stdoutLines = createInterface({ input: child.stdout })
+stdoutLines.on('line', (line) => {
+  if (!printReadyLine(line)) {
+    console.log(line)
+  }
+})
+
+child.once('error', (error) => {
+  console.error(`[headless-pairing] failed to start server: ${error.message}`)
+  cleanupProfile()
+  process.exitCode = 1
+})
+
+child.once('exit', (code, signal) => {
+  stdoutLines.close()
+  if (!sawPairingUrl && code !== 0) {
+    console.error('[headless-pairing] server exited before printing a pairing URL.')
+  }
+  cleanupProfile()
+  if (typeof code === 'number') {
+    process.exitCode = code
+    return
+  }
+  process.exitCode = signal ? 1 : 0
+})
+
+process.on('SIGINT', () => stopChild('SIGINT'))
+process.on('SIGTERM', () => stopChild('SIGTERM'))
+
+/**
+ * Parses wrapper flags and forwards everything else to `orca serve`.
+ */
+function parseArgs(args) {
+  const serveArgs = []
+  let keepProfile = false
+  let help = false
+  for (const arg of args) {
+    if (arg === '--keep') {
+      keepProfile = true
+      continue
+    }
+    if (arg === '-h' || arg === '--help') {
+      help = true
+      continue
+    }
+    serveArgs.push(arg)
+  }
+  return { help, keepProfile, serveArgs }
+}
+
+/**
+ * Prints script usage without touching the dev profile or starting the server.
+ */
+function printHelp() {
+  console.log(`Usage: node config/scripts/serve-headless-fresh-profile-pairing.mjs [--keep] [orca serve flags]
+
+Starts orca-dev serve --json with a fresh isolated userData profile, ensures Electron's dev runtime is usable, and prints the pairing URL.
+
+Wrapper flags:
+  --keep        Keep the fresh profile after the server exits.
+  -h, --help    Show this help.
+
+Forwarded examples:
+  node config/scripts/serve-headless-fresh-profile-pairing.mjs --port 6768
+  node config/scripts/serve-headless-fresh-profile-pairing.mjs --pairing-address 100.64.1.20
+  node config/scripts/serve-headless-fresh-profile-pairing.mjs --mobile-pairing
+
+Environment:
+  ORCA_HEADLESS_PAIRING_ADDRESS=<host|host:port|ws://...>  Override the auto pairing address.
+  ORCA_HEADLESS_PAIRING_PROFILE_DIR=/path/to/profile       Use a fixed profile directory.
+`)
+}
+
+/**
+ * Adds a reachable pairing address unless the caller provided one explicitly.
+ */
+function withDefaultPairingAddress(args) {
+  if (hasPairingAddress(args)) {
+    return args
+  }
+  const address = resolveDefaultPairingAddress()
+  if (!address) {
+    return args
+  }
+  console.error(`[headless-pairing] pairingAddress=${address} (auto)`)
+  return [...args, '--pairing-address', address]
+}
+
+/**
+ * Checks both supported CLI forms for an explicit pairing address.
+ */
+function hasPairingAddress(args) {
+  return hasForwardedServeFlag(args, 'pairing-address')
+}
+
+/**
+ * Checks both exact and `--flag=value` forms for a forwarded serve flag.
+ */
+function hasForwardedServeFlag(args, name) {
+  const flag = `--${name}`
+  return args.some((arg) => arg === flag || arg.startsWith(`${flag}=`))
+}
+
+/**
+ * Prefers an override, then Tailscale, then the OS hostname over loopback.
+ */
+function resolveDefaultPairingAddress() {
+  const configured = process.env.ORCA_HEADLESS_PAIRING_ADDRESS?.trim()
+  if (configured) {
+    return configured
+  }
+  const tailscaleAddress = readTailscaleAddress()
+  if (tailscaleAddress) {
+    return tailscaleAddress
+  }
+  const host = hostname().trim()
+  return host && host !== 'localhost' ? host : null
+}
+
+/**
+ * Reads the first Tailscale IPv4 address when Tailscale is installed.
+ */
+function readTailscaleAddress() {
+  const result = spawnSync('tailscale', ['ip', '-4'], {
+    encoding: 'utf8',
+    stdio: ['ignore', 'pipe', 'ignore']
+  })
+  if (result.error || result.status !== 0) {
+    return null
+  }
+  return (
+    result.stdout
+      .split(/\r?\n/)
+      .map((line) => line.trim())
+      .find(Boolean) ?? null
+  )
+}
+
+/**
+ * Formats forwarded arguments for the startup log.
+ */
+function formatForwardedArgs(args) {
+  if (args.length === 0) {
+    return ''
+  }
+  return ` ${args.map(formatShellArg).join(' ')}`
+}
+
+/**
+ * Quotes a shell-ish argument for display only.
+ */
+function formatShellArg(value) {
+  if (/^[\w./:=@+-]+$/.test(value)) {
+    return value
+  }
+  return JSON.stringify(value)
+}
+
+/**
+ * Runs the same Electron runtime preflight used by dev/start package scripts.
+ */
+function ensureElectronRuntime() {
+  console.error('[headless-pairing] ensuring Electron runtime...')
+  const result = spawnSync(process.execPath, [ensureNativeRuntimeScript, '--runtime=electron'], {
+    cwd: repoRoot,
+    stdio: 'inherit'
+  })
+  if (result.error) {
+    console.error(`[headless-pairing] Electron runtime preflight failed: ${result.error.message}`)
+    process.exit(1)
+  }
+  if (result.status !== 0) {
+    process.exit(result.status ?? 1)
+  }
+}
+
+/**
+ * Rewrites the CLI JSON readiness line into the pairing URL testers need.
+ */
+function printReadyLine(line) {
+  let payload
+  try {
+    payload = JSON.parse(line)
+  } catch {
+    return false
+  }
+  if (!payload || payload.type !== 'orca_server_ready') {
+    return false
+  }
+  console.log(`Orca server ready: ${payload.endpoint ?? 'websocket unavailable'}`)
+  if (payload.pairing?.endpoint) {
+    console.log(`Pairing endpoint: ${payload.pairing.endpoint}`)
+  }
+  if (payload.pairing?.webClientUrl) {
+    console.log(`Web client URL: ${payload.pairing.webClientUrl}`)
+  }
+  if (payload.pairing?.url) {
+    sawPairingUrl = true
+    console.log(`Pairing URL: ${payload.pairing.url}`)
+    console.error('[headless-pairing] server is running; press Ctrl+C to stop.')
+    return true
+  }
+  console.log('Pairing URL: unavailable')
+  return true
+}
+
+/**
+ * Stops the foreground server; repeated signals force its process tree down.
+ */
+function stopChild(signal) {
+  if (!child || child.killed) {
+    return
+  }
+  stopAttempts += 1
+  const targetSignal = stopAttempts > 1 ? 'SIGKILL' : signal
+  if (process.platform === 'win32' && child.pid) {
+    // Why: child.kill() only targets orca-dev on Windows; taskkill walks the
+    // CLI/Electron descendants so the fresh profile is not left locked.
+    const killer = spawn('taskkill', ['/pid', String(child.pid), '/t', '/f'], {
+      stdio: 'ignore',
+      windowsHide: true
+    })
+    killer.on('error', () => child?.kill(targetSignal))
+    return
+  }
+  if (process.platform !== 'win32' && child.pid) {
+    // Why: orca-dev synchronously owns the CLI child, which owns Electron; kill
+    // the spawned process group so programmatic shutdown does not orphan serve.
+    try {
+      process.kill(-child.pid, targetSignal)
+      return
+    } catch {
+      // Fall back to the direct child below if the process group already exited.
+    }
+  }
+  child.kill(targetSignal)
+}
+
+/**
+ * Removes only the ephemeral profile directory this script created.
+ */
+function cleanupProfile() {
+  if (cleanedUp) {
+    return
+  }
+  cleanedUp = true
+  if (parsed.keepProfile || !ownsProfileDir) {
+    console.error(`[headless-pairing] kept ${profileDir}`)
+    return
+  }
+  if (!existsSync(profileDir) || !profileDir.includes('orca-headless-pairing-profile-')) {
+    console.error(`[headless-pairing] skipped cleanup for unexpected profile path: ${profileDir}`)
+    return
+  }
+  try {
+    rmSync(profileDir, { recursive: true, force: true, maxRetries: 5, retryDelay: 200 })
+    console.error(`[headless-pairing] removed ${profileDir}`)
+  } catch (error) {
+    const message = error instanceof Error ? error.message : String(error)
+    console.error(`[headless-pairing] skipped cleanup for ${profileDir}: ${message}`)
+  }
+}

config/scripts/verify-packaged-daemon-entry.cjs

@@ -0,0 +1,46 @@
+const { existsSync } = require('node:fs')
+const { spawnSync } = require('node:child_process')
+const { join } = require('node:path')
+
+// Why: v1.4.129-rc.1 shipped a terminal daemon that could not load (an electron
+// `require` leaked into its bundle) while every build check passed. This boots
+// the PACKAGED daemon-entry under plain Node against the asar-unpacked layout,
+// so a bundling / asar-unpack regression fails packaging instead of reaching
+// users. Module-load proof only: with no args the entry must reach argv parsing
+// and print its "Usage: daemon-entry" error — a MODULE_NOT_FOUND or a missing
+// usage line means the packaged graph does not load and the build must fail.
+//
+// resourcesDir is the packaged Resources dir (Contents/Resources on macOS,
+// <appOutDir>/resources elsewhere). execPath defaults to the packaging Node.
+function verifyPackagedDaemonEntryBoots(resourcesDir, options = {}) {
+  const execPath = options.execPath || process.execPath
+  const entryPath = join(resourcesDir, 'app.asar.unpacked', 'out', 'main', 'daemon-entry.js')
+  if (!existsSync(entryPath)) {
+    // Why: some targets/layouts do not unpack here; skip rather than fail so
+    // the hook stays safe across platforms it has not verified.
+    console.log(`[verify-packaged-daemon-entry] skipped — no unpacked entry at ${entryPath}`)
+    return
+  }
+
+  const result = spawnSync(execPath, [entryPath], { encoding: 'utf8', timeout: 10_000 })
+  if (result.error) {
+    throw new Error(
+      `[verify-packaged-daemon-entry] could not launch daemon-entry.js: ${result.error.message}`
+    )
+  }
+  const stderr = result.stderr || ''
+  if (/Cannot find module|MODULE_NOT_FOUND/.test(stderr)) {
+    throw new Error(
+      `[verify-packaged-daemon-entry] packaged daemon-entry.js failed to load under plain Node:\n${stderr}`
+    )
+  }
+  if (!stderr.includes('Usage: daemon-entry')) {
+    throw new Error(
+      `[verify-packaged-daemon-entry] packaged daemon-entry.js did not reach argv parsing ` +
+        `(expected the "Usage: daemon-entry" error). stderr:\n${stderr}`
+    )
+  }
+  console.log('[verify-packaged-daemon-entry] OK — packaged daemon-entry loads under plain Node')
+}
+
+module.exports = { verifyPackagedDaemonEntryBoots }

config/scripts/verify-release-required-assets.mjs

@@ -39,9 +39,9 @@ export function extractManifestAssetNames(manifestText) {
     }
     const value = match[1].trim()
     try {
-      names.add(new URL(value).pathname.split('/').filter(Boolean).at(-1) ?? value)
+      names.add(new URL(value).pathname.split('/').findLast(Boolean) ?? value)
     } catch {
-      names.add(value.split('/').filter(Boolean).at(-1) ?? value)
+      names.add(value.split('/').findLast(Boolean) ?? value)
     }
   }
   return [...names]

config/tsconfig.cli.json

@@ -13,9 +13,12 @@
     "../src/main/codex/codex-config-mirror.ts",
     "../src/main/codex/codex-config-path-reference-rewrite.ts",
     "../src/main/codex/codex-home-paths.ts",
+    "../src/main/codex/codex-hook-identity.ts",
+    "../src/main/codex/config-settings-promotion.ts",
     "../src/main/codex/config-toml-line-scan.ts",
     "../src/main/codex/config-toml-trust.ts",
     "../src/main/codex/hook-service.ts",
+    "../src/main/codex/hook-trust-promotion.ts",
     "../src/main/codex-accounts/fs-utils.ts",
     "../src/main/command-code/command-code-managed-script.ts",
     "../src/main/command-code/hook-service.ts",

config/tsconfig.node.json

@@ -2,6 +2,7 @@
   "extends": "@electron-toolkit/tsconfig/tsconfig.node.json",
   "include": [
     "../electron.vite.config.*",
+    "../build-plugins/**/*",
     "../src/main/**/*",
     "../src/preload/**/*",
     "../src/shared/**/*",

docs/assets/readme-downloads.svg

@@ -1,5 +1,5 @@
-<svg xmlns="http://www.w3.org/2000/svg" width="113" height="20" role="img" aria-label="downloads: 3.4m">
-  <title>downloads: 3.4m</title>
+<svg xmlns="http://www.w3.org/2000/svg" width="113" height="20" role="img" aria-label="downloads: 4.1m">
+  <title>downloads: 4.1m</title>
   <linearGradient id="s" x2="0" y2="100%">
     <stop offset="0" stop-color="#bbb" stop-opacity=".1"/>
     <stop offset="1" stop-opacity=".1"/>
@@ -15,7 +15,7 @@
   <g fill="#fff" text-anchor="middle" font-family="Verdana,Geneva,DejaVu Sans,sans-serif" text-rendering="geometricPrecision" font-size="11">
     <text x="37" y="15" fill="#010101" fill-opacity=".3">downloads</text>
     <text x="37" y="14">downloads</text>
-    <text x="93.5" y="15" fill="#010101" fill-opacity=".3">3.4m</text>
-    <text x="93.5" y="14">3.4m</text>
+    <text x="93.5" y="15" fill="#010101" fill-opacity=".3">4.1m</text>
+    <text x="93.5" y="14">4.1m</text>
   </g>
 </svg>

docs/readme/README.es.md

@@ -12,7 +12,7 @@
 </p>
 
 <p align="center">
-  <sub><a href="../../README.md">English</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a></sub>
+  <sub><a href="../../README.md">English</a> · <a href="README.pt.md">Português</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a></sub>
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@
 
 Supervisa y dirige a tus agentes desde el teléfono — recibe una notificación cuando un agente termine y envía instrucciones de seguimiento desde cualquier lugar.
 
-[App Store de iOS](https://apps.apple.com/us/app/orca-ide/id6766130217) · [APK para Android](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk) · [Docs →](https://www.onorca.dev/docs/mobile)
+[App Store de iOS](https://apps.apple.com/us/app/orca-ide/id6766130217) · [APK para Android](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [Docs →](https://www.onorca.dev/docs/mobile)
 
 </td>
 <td width="50%">
@@ -227,7 +227,7 @@ yay -S stably-orca-bin
 Vincúlala con tu app de escritorio para supervisar y dirigir a tus agentes desde el teléfono.
 
 - **iOS:** [Descargar desde App Store](https://apps.apple.com/us/app/orca-ide/id6766130217)
-- **Android:** [Descargar el APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk)
+- **Android:** [Descargar el APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
 
 ---
 

docs/readme/README.ja.md

@@ -12,7 +12,7 @@
 </p>
 
 <p align="center">
-  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ko.md">한국어</a></sub>
+  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.pt.md">Português</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ko.md">한국어</a></sub>
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@
 
 スマートフォンからエージェントを監視・操作 — エージェントの完了を通知で受け取り、どこからでもフォローアップを送信できます。
 
-[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk) · [ドキュメント →](https://www.onorca.dev/docs/mobile)
+[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [ドキュメント →](https://www.onorca.dev/docs/mobile)
 
 </td>
 <td width="50%">
@@ -227,7 +227,7 @@ yay -S stably-orca-bin
 デスクトップアプリとペアリングして、スマートフォンからエージェントを監視・操作できます。
 
 - **iOS:** [App Store からダウンロード](https://apps.apple.com/us/app/orca-ide/id6766130217)
-- **Android:** [APK をダウンロード](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk)
+- **Android:** [APK をダウンロード](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
 
 ---
 

docs/readme/README.ko.md

@@ -12,7 +12,7 @@
 </p>
 
 <p align="center">
-  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a></sub>
+  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.pt.md">Português</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a></sub>
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@
 
 휴대폰에서 에이전트를 모니터링하고 조종하세요 — 에이전트가 완료되면 알림을 받고 어디서든 후속 지시를 보낼 수 있습니다.
 
-[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk) · [문서 →](https://www.onorca.dev/docs/mobile)
+[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [문서 →](https://www.onorca.dev/docs/mobile)
 
 </td>
 <td width="50%">
@@ -227,7 +227,7 @@ yay -S stably-orca-bin
 데스크톱 앱과 페어링해 휴대폰에서 에이전트를 모니터링하고 조종하세요.
 
 - **iOS:** [App Store에서 다운로드](https://apps.apple.com/us/app/orca-ide/id6766130217)
-- **Android:** [APK 다운로드](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk)
+- **Android:** [APK 다운로드](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
 
 ---
 

docs/readme/README.pt.md

@@ -0,0 +1,257 @@
+<h1 align="center">
+  <a href="https://onOrca.dev"><img src="../../resources/build/icon.png" alt="Orca" width="64" valign="middle" /></a> Orca
+</h1>
+
+<p align="center">
+  <a href="https://github.com/stablyai/orca/stargazers"><img src="https://badgen.net/github/stars/stablyai/orca?label=%E2%98%85" alt="Estrelas no GitHub" /></a>
+  <a href="https://github.com/stablyai/orca/releases"><img src="../assets/readme-downloads.svg" alt="Total de downloads em todas as versões" /></a>
+  <img src="https://badgen.net/github/license/stablyai/orca" alt="Licença" />
+  <a href="https://discord.gg/fzjDKHxv8Q"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Entre no Discord do Orca" /></a>
+  <a href="https://x.com/orca_build"><img src="https://img.shields.io/badge/X-000000?logo=x&logoColor=white" alt="Siga o Orca no X" /></a>
+  <img src="https://img.shields.io/badge/macOS%20%7C%20Windows%20%7C%20Linux-4493F8?style=flat-square" alt="Plataformas compatíveis: macOS, Windows e Linux" />
+</p>
+
+<p align="center">
+  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.zh-CN.md">中文</a> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a></sub>
+</p>
+
+<p align="center">
+  <strong>O orquestrador de IA para builders 100x.</strong><br/>
+  Rode Codex, ClaudeCode, OpenCode ou Pi lado a lado — cada um em seu próprio worktree, acompanhado em um só lugar.
+</p>
+
+<h3 align="center"><a href="https://onorca.dev/download"><ins>Baixar o Orca</ins></a></h3>
+
+<p align="center">
+  <img src="../assets/readme-hero.jpg" alt="App desktop do Orca executando agentes em worktrees paralelos, com o app companion móvel do Orca no canto" width="960" />
+</p>
+
+## Recursos
+
+<table>
+<tr>
+<td width="50%" valign="middle">
+
+### Companion móvel
+
+Monitore e conduza seus agentes pelo celular — receba uma notificação quando um agente terminar e envie instruções de acompanhamento de qualquer lugar.
+
+[App Store para iOS](https://apps.apple.com/us/app/orca-ide/id6766130217) · [TestFlight](https://testflight.apple.com/join/YjeGMQBA) · [APK Android 0.0.25](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [Docs →](https://www.onorca.dev/docs/mobile)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/mobile"><picture><source srcset="../assets/feature-wall/mobile-companion-app-showcase.gif" type="image/gif"><img src="../assets/feature-wall/mobile-companion-app-showcase.jpg" alt="Orca desktop com o app companion móvel" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Worktrees paralelos
+
+Envie um mesmo prompt para cinco agentes, cada um em seu próprio worktree git isolado — compare os resultados e faça merge do vencedor.
+
+[Docs →](https://www.onorca.dev/docs/model/worktrees)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/model/worktrees"><picture><source srcset="../assets/feature-wall/parallel-worktrees.gif" type="image/gif"><img src="../assets/feature-wall/parallel-worktrees.jpg" alt="Orquestração de worktrees paralelos" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Terminais divididos
+
+Terminais no nível do Ghostty com renderização WebGL, divisões infinitas e scrollback que sobrevive a reinicializações.
+
+[Docs →](https://www.onorca.dev/docs/terminal)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/terminal"><picture><source srcset="../assets/feature-wall/terminal-splits.gif" type="image/gif"><img src="../assets/feature-wall/terminal-splits.jpg" alt="Terminais divididos" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Modo Design
+
+Clique em qualquer elemento de UI em uma janela real do Chromium para enviar HTML, CSS e uma captura recortada direto para o prompt do seu agente.
+
+[Docs →](https://www.onorca.dev/docs/browser/design-mode)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/browser/design-mode"><picture><source srcset="../assets/feature-wall/design-mode.gif" type="image/gif"><img src="../assets/feature-wall/design-mode.jpg" alt="Navegador integrado e Modo Design" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### GitHub e Linear nativos
+
+Navegue por PRs, issues e quadros de projeto dentro do app — abra um worktree a partir de qualquer tarefa e revise sem trocar de contexto.
+
+[Docs →](https://www.onorca.dev/docs/review/linear)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/review/linear"><picture><source srcset="../assets/feature-wall/github-linear.gif" type="image/gif"><img src="../assets/feature-wall/github-linear.jpg" alt="Fluxos de trabalho de tarefas do GitHub e Linear no Orca" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Worktrees por SSH
+
+Execute agentes em uma máquina remota potente com edição completa de arquivos, git e terminais — com reconexão automática e encaminhamento de portas incluídos.
+
+[Docs →](https://www.onorca.dev/docs/ssh)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/ssh"><picture><source srcset="../assets/feature-wall/ssh-worktrees.gif" type="image/gif"><img src="../assets/feature-wall/ssh-worktrees.jpg" alt="Worktrees remotos por SSH" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Anotar diffs de IA
+
+Deixe comentários em qualquer linha de diff e envie-os de volta ao agente — revise, edite e faça commit sem sair do Orca.
+
+[Docs →](https://www.onorca.dev/docs/review/annotate-ai-diff)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/review/annotate-ai-diff"><picture><source srcset="../assets/feature-wall/annotate-diff.gif" type="image/gif"><img src="../assets/feature-wall/annotate-diff.jpg" alt="Anotar diffs gerados por IA" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Arraste arquivos para agentes
+
+O editor do VS Code com salvamento automático em todos os lugares — arraste arquivos ou imagens direto para o prompt de um agente.
+
+[Docs →](https://www.onorca.dev/docs/editing/file-explorer)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/editing/file-explorer"><picture><source srcset="../assets/feature-wall/file-drag.gif" type="image/gif"><img src="../assets/feature-wall/file-drag.jpg" alt="Arraste arquivos e imagens para o prompt de um agente" width="100%" /></picture></a>
+</td>
+</tr>
+<tr>
+<td width="50%" valign="middle">
+
+### Orca CLI
+
+Agentes também controlam o Orca — automatize qualquer fluxo de trabalho com `orca worktree create`, `snapshot`, `click` e `fill`.
+
+[Docs →](https://www.onorca.dev/docs/cli/overview)
+
+</td>
+<td width="50%">
+  <a href="https://www.onorca.dev/docs/cli/overview"><picture><source srcset="../assets/feature-wall/orca-cli.gif" type="image/gif"><img src="../assets/feature-wall/orca-cli.jpg" alt="Automatize o Orca pela CLI" width="100%" /></picture></a>
+</td>
+</tr>
+</table>
+
+**Também incluído:**
+
+- **[Abertura rápida](https://www.onorca.dev/docs/model/quick-open)** — Pesquise entre worktrees, arquivos, agentes, comandos e contexto do repositório sem sair do seu fluxo.
+- **[Troca de contas e acompanhamento de uso](https://www.onorca.dev/docs/agents/usage-tracking)** — Veja o uso de Claude e Codex, os reinícios de limites e troque contas instantaneamente sem fazer login de novo.
+- **[Prévias ricas do repositório](https://www.onorca.dev/docs/editing/markdown)** — Pré-visualize Markdown, imagens, PDFs e documentos do repositório no workspace.
+- **[Computer Use](https://www.onorca.dev/docs/cli/computer-use)** — Deixe agentes operarem apps de desktop e UI visível quando um fluxo de trabalho precisa de interação real.
+- **[Notificações e estado de não lido](https://www.onorca.dev/docs/notifications)** — Saiba quando um agente termina ou precisa de atenção, depois marque conversas como não lidas para voltar depois.
+- **E muito, muito mais** — lançamos novidades todos os dias, então esta lista vive atrasada. O [changelog](https://github.com/stablyai/orca/releases) é a lista real de recursos.
+
+---
+
+## Agentes compatíveis
+
+Funciona com **qualquer agente CLI** — se roda em um terminal, roda no Orca.
+
+<p>
+  <a href="https://docs.anthropic.com/claude/docs/claude-code"><kbd><img src="../assets/claude-logo.svg" alt="Logotipo do Claude Code" width="16" valign="middle" /> Claude Code</kbd></a> &nbsp;
+  <a href="https://github.com/openai/codex"><kbd><img src="https://www.google.com/s2/favicons?domain=openai.com&sz=64" alt="Logotipo do Codex" width="16" valign="middle" /> Codex</kbd></a> &nbsp;
+  <a href="https://x.ai/cli"><kbd><img src="https://www.google.com/s2/favicons?domain=x.ai&sz=64" alt="Logotipo do Grok" width="16" valign="middle" /> Grok</kbd></a> &nbsp;
+  <a href="https://cursor.com/cli"><kbd><img src="https://www.google.com/s2/favicons?domain=cursor.com&sz=64" alt="Logotipo do Cursor" width="16" valign="middle" /> Cursor</kbd></a> &nbsp;
+  <a href="https://docs.github.com/en/copilot/how-tos/set-up/install-copilot-cli"><kbd><img src="https://www.google.com/s2/favicons?domain=github.com&sz=64" alt="Logotipo do GitHub Copilot" width="16" valign="middle" /> GitHub Copilot</kbd></a> &nbsp;
+  <a href="https://opencode.ai/docs/cli/"><kbd><img src="https://www.google.com/s2/favicons?domain=opencode.ai&sz=64" alt="Logotipo do OpenCode" width="16" valign="middle" /> OpenCode</kbd></a> &nbsp;
+  <a href="https://mimo.xiaomi.com/coder"><kbd><img src="https://www.google.com/s2/favicons?domain=mimo.xiaomi.com&sz=64" alt="Logotipo do MiMo Code" width="16" valign="middle" /> MiMo Code</kbd></a> &nbsp;
+  <a href="https://ampcode.com/manual#install"><kbd><img src="https://www.google.com/s2/favicons?domain=ampcode.com&sz=64" alt="Logotipo do Amp" width="16" valign="middle" /> Amp</kbd></a> &nbsp;
+  <a href="https://openclaude.gitlawb.com/"><kbd><img src="../../resources/openclaude-logo.png" alt="Logotipo do OpenClaude" width="16" valign="middle" /> OpenClaude</kbd></a> &nbsp;
+  <a href="https://antigravity.google/docs/cli-overview"><kbd><img src="https://www.google.com/s2/favicons?domain=antigravity.google&sz=64" alt="Logotipo do Antigravity" width="16" valign="middle" /> Antigravity</kbd></a> &nbsp;
+  <a href="https://pi.dev"><kbd><img src="https://pi.dev/favicon.svg" alt="Logotipo do Pi" width="16" valign="middle" /> Pi</kbd></a> &nbsp;
+  <a href="https://omp.sh"><kbd><img src="https://omp.sh/favicon.svg" alt="Logotipo do oh-my-pi" width="16" valign="middle" /> oh-my-pi</kbd></a> &nbsp;
+  <a href="https://hermes-agent.nousresearch.com/docs/"><kbd><img src="https://www.google.com/s2/favicons?domain=nousresearch.com&sz=64" alt="Logotipo do Hermes Agent" width="16" valign="middle" /> Hermes Agent</kbd></a> &nbsp;
+  <a href="https://devin.ai/cli"><kbd><img src="https://www.google.com/s2/favicons?domain=devin.ai&sz=64" alt="Logotipo do Devin" width="16" valign="middle" /> Devin</kbd></a> &nbsp;
+  <a href="https://block.github.io/goose/docs/quickstart/"><kbd><img src="https://www.google.com/s2/favicons?domain=goose-docs.ai&sz=64" alt="Logotipo do Goose" width="16" valign="middle" /> Goose</kbd></a> &nbsp;
+  <a href="https://docs.augmentcode.com/cli/overview"><kbd><img src="https://www.google.com/s2/favicons?domain=augmentcode.com&sz=64" alt="Logotipo do Auggie" width="16" valign="middle" /> Auggie</kbd></a> &nbsp;
+  <a href="https://github.com/autohandai/code-cli"><kbd><img src="https://www.google.com/s2/favicons?domain=autohand.ai&sz=64" alt="Logotipo do Autohand Code" width="16" valign="middle" /> Autohand Code</kbd></a> &nbsp;
+  <a href="https://github.com/charmbracelet/crush"><kbd><img src="https://www.google.com/s2/favicons?domain=charm.sh&sz=64" alt="Logotipo do Charm" width="16" valign="middle" /> Charm</kbd></a> &nbsp;
+  <a href="https://docs.cline.bot/cline-cli/overview"><kbd><img src="https://www.google.com/s2/favicons?domain=cline.bot&sz=64" alt="Logotipo do Cline" width="16" valign="middle" /> Cline</kbd></a> &nbsp;
+  <a href="https://www.codebuff.com/docs/help/quick-start"><kbd><img src="https://www.google.com/s2/favicons?domain=codebuff.com&sz=64" alt="Logotipo do Codebuff" width="16" valign="middle" /> Codebuff</kbd></a> &nbsp;
+  <a href="https://commandcode.ai/docs/quickstart"><kbd><img src="https://www.google.com/s2/favicons?domain=commandcode.ai&sz=64" alt="Logotipo do Command Code" width="16" valign="middle" /> Command Code</kbd></a> &nbsp;
+  <a href="https://docs.continue.dev/guides/cli"><kbd><img src="https://www.google.com/s2/favicons?domain=continue.dev&sz=64" alt="Logotipo do Continue" width="16" valign="middle" /> Continue</kbd></a> &nbsp;
+  <a href="https://docs.factory.ai/cli/getting-started/quickstart"><kbd><img src="../assets/droid-logo.svg" alt="Logotipo do Droid" width="16" valign="middle" /> Droid</kbd></a> &nbsp;
+  <a href="https://kilo.ai/docs/cli"><kbd><img src="https://raw.githubusercontent.com/Kilo-Org/kilocode/main/packages/kilo-vscode/assets/icons/kilo-light.svg" alt="Logotipo do Kilocode" width="16" valign="middle" /> Kilocode</kbd></a> &nbsp;
+  <a href="https://www.kimi.com/code/docs/en/kimi-code-cli/getting-started.html"><kbd><img src="https://www.google.com/s2/favicons?domain=moonshot.cn&sz=64" alt="Logotipo do Kimi" width="16" valign="middle" /> Kimi</kbd></a> &nbsp;
+  <a href="https://kiro.dev/docs/cli/"><kbd><img src="https://www.google.com/s2/favicons?domain=kiro.dev&sz=64" alt="Logotipo do Kiro" width="16" valign="middle" /> Kiro</kbd></a> &nbsp;
+  <a href="https://github.com/mistralai/mistral-vibe"><kbd><img src="https://www.google.com/s2/favicons?domain=mistral.ai&sz=64" alt="Logotipo do Mistral Vibe" width="16" valign="middle" /> Mistral Vibe</kbd></a> &nbsp;
+  <a href="https://github.com/QwenLM/qwen-code"><kbd><img src="https://www.google.com/s2/favicons?domain=qwenlm.github.io&sz=64" alt="Logotipo do Qwen Code" width="16" valign="middle" /> Qwen Code</kbd></a> &nbsp;
+  <a href="https://support.atlassian.com/rovo/docs/install-and-run-rovo-dev-cli-on-your-device/"><kbd><img src="https://www.google.com/s2/favicons?domain=atlassian.com&sz=64" alt="Logotipo do Rovo Dev" width="16" valign="middle" /> Rovo Dev</kbd></a> &nbsp;
+  <kbd>+ qualquer agente CLI</kbd>
+</p>
+
+---
+
+## Instalação
+
+### Desktop — macOS, Windows, Linux
+
+- **[Baixe em onOrca.dev](https://onorca.dev/download)**
+- Ou baixe um build diretamente: [macOS Apple Silicon](https://github.com/stablyai/orca/releases/latest/download/orca-macos-arm64.dmg) · [macOS Intel](https://github.com/stablyai/orca/releases/latest/download/orca-macos-x64.dmg) · [Windows (.exe)](https://github.com/stablyai/orca/releases/latest/download/orca-windows-setup.exe) · [Linux AppImage](https://github.com/stablyai/orca/releases/latest/download/orca-linux.AppImage) · [Todos os builds](https://github.com/stablyai/orca/releases/latest)
+- Rodando `orca serve` em um servidor Linux headless? Veja o [guia de servidor Linux headless](../reference/headless-linux-server.md).
+
+_Ou por um gerenciador de pacotes:_
+
+```bash
+# macOS (Homebrew)
+brew install --cask stablyai/orca/orca
+
+# Arch Linux (AUR) — ou stably-orca-git para compilar a partir do código-fonte
+yay -S stably-orca-bin
+```
+
+### Companion móvel — iOS, Android
+
+Conecte ao app desktop para monitorar e conduzir seus agentes pelo celular.
+
+- **iOS:** [Baixar na App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) ou [entrar no TestFlight](https://testflight.apple.com/join/YjeGMQBA)
+- **Android:** [Baixar APK 0.0.25](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
+
+---
+
+## Comunidade e suporte
+
+- **Discord:** Entre na comunidade no **[Discord](https://discord.gg/fzjDKHxv8Q)**.
+- **Twitter / X:** Siga **[@orca_build](https://x.com/orca_build)** para atualizações e anúncios.
+- **Feedback e ideias:** Lançamos rápido. Sentiu falta de algo? [Peça um novo recurso](https://github.com/stablyai/orca/issues).
+- **Privacidade:** Veja a [documentação de privacidade e telemetria](https://www.onorca.dev/docs/telemetry) para saber quais dados anônimos de uso o Orca coleta e como desativar.
+- **Mostre apoio:** Dê uma [estrela](https://github.com/stablyai/orca) neste repositório para acompanhar nossos lançamentos diários.
+
+---
+
+## Desenvolvimento
+
+Quer contribuir ou rodar localmente? Veja nosso guia [CONTRIBUTING.md](../../.github/CONTRIBUTING.md).
+
+<a href="https://github.com/stablyai/orca/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=stablyai/orca" alt="Contribuidores do Orca" />
+</a>
+
+## Licença
+
+Orca é livre e de código aberto sob a [Licença MIT](../../LICENSE).

docs/readme/README.zh-CN.md

@@ -12,7 +12,7 @@
 </p>
 
 <p align="center">
-  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a></sub>
+  <sub><a href="../../README.md">English</a> · <a href="README.es.md">Español</a> · <a href="README.pt.md">Português</a> · <a href="README.ja.md">日本語</a> · <a href="README.ko.md">한국어</a></sub>
 </p>
 
 <p align="center">
@@ -36,7 +36,7 @@
 
 用手机监控并指挥你的智能体 — 智能体完成时收到通知，随时随地发送后续指令。
 
-[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk) · [文档 →](https://www.onorca.dev/docs/mobile)
+[iOS App Store](https://apps.apple.com/us/app/orca-ide/id6766130217) · [Android APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk) · [文档 →](https://www.onorca.dev/docs/mobile)
 
 </td>
 <td width="50%">
@@ -227,7 +227,7 @@ yay -S stably-orca-bin
 与桌面应用配对，用手机监控并指挥你的智能体。
 
 - **iOS:** [从 App Store 下载](https://apps.apple.com/us/app/orca-ide/id6766130217)
-- **Android:** [下载 APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.21/app-release.apk)
+- **Android:** [下载 APK](https://github.com/stablyai/orca/releases/download/mobile-android-v0.0.25/app-release.apk)
 
 ---
 
@@ -235,6 +235,11 @@ yay -S stably-orca-bin
 
 - **Discord:** 加入 **[Discord](https://discord.gg/fzjDKHxv8Q)** 社区。
 - **Twitter / X:** 关注 **[@orca_build](https://x.com/orca_build)** 获取更新和公告。
+- **微信:** 扫描二维码加入社区。如果第一个群已满，请使用备用微信群。
+
+  <img src="../assets/wechat-qr.png" alt="Orca 社区微信群二维码" width="160" />
+  <img src="../assets/wechat-qr-backup.jpg" alt="Orca 社区备用微信群二维码" width="160" />
+
 - **反馈与想法:** 我们发布很快。缺少什么功能？[提交功能请求](https://github.com/stablyai/orca/issues)。
 - **隐私:** 查看[隐私与遥测文档](https://www.onorca.dev/docs/telemetry)，了解 Orca 收集哪些匿名使用数据以及如何退出。
 - **支持我们:** 给这个仓库点 [Star](https://github.com/stablyai/orca)，关注我们的日常发布。

docs/reference/reliability-gates-implementation-plan.md

@@ -0,0 +1,783 @@
+# Reliability Gates Implementation Plan
+
+Date: 2026-07-02
+
+Source context: [`docs/reference/reliability-pain-points-2026-06-30.md`](./reliability-pain-points-2026-06-30.md).
+
+## Current Status
+
+Last updated: 2026-07-03.
+
+The first repo-side reliability batch is useful, but it should be treated as a first-layer regression net, not a finished terminal reliability program. Before final merge claims, rebase or retarget the stack onto fresh `origin/main` and verify it still includes the protections that landed there after the branch was cut.
+
+Stack topology facts the rest of this plan depends on (verified 2026-07-02):
+
+- `brennanb2025/fix-terminal-reliability` is simultaneously the integration worktree and the head branch of #7001, and it is the base branch of all five child PRs (#7004-#7008). Committing broad product changes to it would turn #7001 into the single large terminal PR this plan forbids and would silently change every child PR's diff.
+- The committed branch contains only the manifest, checker, checker tests, and these docs (a 5-gate manifest as committed). Everything else this plan describes — the expanded manifest, product hardening, and new or updated tests across roughly 39 modified files plus 3 untracked files — exists only as uncommitted working-tree changes in this worktree. No evidence run below is reproducible from any pushed ref until the split branches are committed and pushed.
+- The branch's merge-base with `origin/main` is roughly 97 commits stale. Fresh main renamed `terminal-webgl-paste-recovery.{ts,test.ts}` to `terminal-webgl-atlas-recovery.{ts,test.ts}` in #6949, so the current `xterm-addon.boundary-containment` manifest entry and its documented command break on rebase; see Fresh-Eyes Corrections.
+- The manifest is the single source of truth for gate counts and per-gate status. Where this document records counts they are dated snapshots; `pnpm run check:reliability-gates` prints the current gate total.
+
+Known Orca PR stack:
+
+- [#7001](https://github.com/stablyai/orca/pull/7001), `Add reliability gate manifest`: merged on 2026-07-03. The manifest, checker, and plan docs are on main; [#7295](https://github.com/stablyai/orca/pull/7295) hardens the checker policy and registers the merged #7133/#7148/#7173/#7192 regression tests as partial gates with fresh main-side evidence.
+- [#7005](https://github.com/stablyai/orca/pull/7005), `Add terminal snapshot freshness contract gate`: depends on #7001 and protects the stale-liveness/newborn-PTY class behind `terminal-session.snapshot-freshness`.
+- [#7008](https://github.com/stablyai/orca/pull/7008), `Expand provider session replay ownership gate`: depends on #7001 and protects the agent/provider ownership class behind `agent-session.provider-ownership`.
+- [#7004](https://github.com/stablyai/orca/pull/7004), `Contain xterm addon failures to terminal panes`: depends on #7001. The current branch has a first executable xterm containment slice for addon load, link-provider, search, and WebGL failure containment; it still needs live typed input/output survival before promotion.
+- [#7006](https://github.com/stablyai/orca/pull/7006), `Prove visible terminal size convergence`: depends on #7001. The current branch now has a first deterministic renderer/main contract slice for post-spawn reconcile, split-right 0x0 recovery, applied `pty:getSize`, and resume-time size reassertion; it still needs live shell, SSH/remote, and Windows ConPTY geometry proof before promotion.
+- [#7007](https://github.com/stablyai/orca/pull/7007), `Add persisted session upgrade fixture gate`: depends on #7001. The current manifest entry is still commandless in this branch, so treat it as a registered startup-upgrade gap until immutable fixture commands are wired.
+
+Merged related product hardening:
+
+- [#7054](https://github.com/stablyai/orca/pull/7054), `Prevent hidden terminal TUI overlap`: merged on 2026-07-01 with `verify`, Wayland terminal input, golden Linux E2E, golden mac E2E, and community tracking checks passing. This fixes the hidden regular-terminal TUI overlap class by scheduling bounded post-parse WebGL atlas recovery for risky hidden TUI redraws while keeping hidden bytes on the background write path.
+- [#7133](https://github.com/stablyai/orca/pull/7133), `Fix stale terminal frames on worktree return`: merged on 2026-07-02 as a high-priority regression fix, root-caused on a live repro. Hidden-output restore skipped the destructive clear for alternate-screen snapshots, `?1049h` is a no-op on a pane already on the alt screen, and serialized frames skip blank cells, so the pre-hide frame bled through the final frame until a select or resize. The fix writes an alt-only `\x1b[?1049h\x1b[2J\x1b[H` preamble before alt-screen snapshot data (normal-buffer scrollback untouched, still no `3J`) and hardens the WebGL reveal path: dispose-on-bail with a single-addon invariant, refresh-always atlas reset, and a settled-frame pane-scoped repaint on tab reveal, worktree resume, and window wake. It supersedes the reverted #7058 approach and closes the alt-screen hole that #7054's hidden-output skip machinery made load-bearing. Its deterministic tests — the alt-only-clear-before-snapshot assertions in `pty-connection.test.ts`, plus `pane-webgl-renderer.test.ts`, `pane-reveal-repaint.test.ts`, and `terminal-visibility-resume.test.ts` — are the seed assertions for `terminal-output.scrollback-restore` and the WebGL side of `xterm-addon.boundary-containment` once the stack rebases.
+
+- [#7148](https://github.com/stablyai/orca/pull/7148), `Match main terminal mirror character widths to the renderer`: merged on 2026-07-02, hours after #7133. Root cause of the residual post-#7133 bottom-row tears: the renderer xterm measures with Unicode 11 + ZWJ joining while the daemon headless mirror used default-width tables, so every positioned write after an emoji lands shifted and the tears are baked into the very mirror that restores replay — #7133's clear made restores faithful to the mirror and exposed it. The fix moves the unicode provider to `src/shared/terminal-unicode-provider.ts` and activates it in the headless emulator. Its red-green width test, `src/main/daemon/headless-emulator-unicode-width.test.ts` on main, is the seed assertion for the new `terminal-mirror.parser-parity` gate, and the shared provider plus `pane-terminal-options.ts` are the shared-construction seams that gate must assert on. #7148 also touches `pane-lifecycle.ts`, which both this worktree and #7004's branch modify — another confirmed rebase-conflict site.
+- [#7173](https://github.com/stablyai/orca/pull/7173), `Fix Codex pane output tearing after app occlusion`: merged on 2026-07-03. A delayed background-origin Codex chunk carrying a stateful query could take the hidden live-write path after a newer snapshot restore had already rebuilt the renderer, overwriting newer state; an adversarial review pass added seq-restart guards so revived sessions (hibernation wake, kill plus cold-restore under the same PTY id) cannot silently freeze hidden output against a stale seq high-water mark. Its three red-proven tests in `pty-connection.test.ts` on main — reference-parse buffer equality over the ordered raw stream plus both revival guards — are exactly the exactness-oracle shape this plan prescribes and seed the hidden/live interleaving slice of `terminal-output.scrollback-restore`.
+- [#7192](https://github.com/stablyai/orca/pull/7192), `Keep runtime mirror sized with desktop terminal resizes`: merged on 2026-07-03. Desktop `pty:resize` updated the PTY and renderer but never the runtime mirror, so the mirror parsed all subsequent bytes at its birth width, soft-wrapped TUI repaints into a corrupted ladder, and baked that corruption into every restore snapshot; the fix fans accepted desktop resizes to the mirror and serializes mirror reflow with queued output writes on the per-PTY write chain. This exposed a plan gap now fixed: the runtime mirror is a geometry authority in its own right, and renderer-vs-mirror convergence checks cannot see corruption upstream of the mirror. Its red-proven tests in `orca-runtime.test.ts` and `pty.test.ts` on main seed the mirror-geometry slice of `terminal-geometry.visible-convergence`.
+
+Factory and skill status:
+
+- [orca-internal #491](https://github.com/stablyai/orca-internal/pull/491), `Require reliability gates in factory reviews`: merged on 2026-07-01. This updates the factory review flow so terminal/session/provider/startup/release changes must name the relevant reliability class, invariant, gate or accepted gap, and validation evidence.
+- [orca-internal #493](https://github.com/stablyai/orca-internal/pull/493), `Add Playwright reliability test skill`: as last reviewed, this still needed rebase/regeneration before merge because the reliability plan depends on Playwright tests being evidence-based, low-flake, and tied to named invariants.
+
+Remaining work:
+
+- Rebase or retarget the reliability stack on fresh `origin/main` before making final merge claims. Current main already contains terminal fixes that the stack must absorb, including hidden TUI overlap, input-lag, scrollback replay, Windows ConPTY keyboard reset, resume width flicker, remote mirror polling, and the #7133 stale alt-screen frame fixes. As part of the rebase, mechanically verify every declared `testFiles` entry in the manifest exists on fresh `origin/main`; the #6949 WebGL rename is a confirmed break. #7133 (merged 2026-07-02) touches `pty-connection.{ts,test.ts}` and `use-terminal-pane-global-effects.test.ts`, which this worktree also modifies, and `pane-webgl-renderer.{ts,test.ts}`, which #7004's branch also edits — expect real conflicts there. #7148 (merged 2026-07-02) additionally touches `pane-lifecycle.ts` and `headless-emulator.ts`, so the worktree's and #7004's pane-lifecycle edits must absorb its shared unicode-provider wiring. #7173 (merged 2026-07-03) rewrites the hidden live-write path in `pty-connection.{ts,test.ts}` and #7192 (merged 2026-07-03) touches `pty.test.ts` — both carry heavy worktree modifications, so the conflict surface keeps widening the longer the rebase waits. This worktree's replay/SSH/backpressure changes and #7004's containment changes must be rebased over #7133's alt-clear preamble, #7148's width fix, #7173's seq filtering, #7192's mirror resize fan-out, and the reveal-repaint hardening without weakening any side.
+- Materialize the split: commit each uncommitted working-tree slice to its owning branch per the file assignment in the PR Split Plan, reset diverged child branches from the worktree versions of their files, and push. Only manifest, checker, and doc changes may be committed to `brennanb2025/fix-terminal-reliability` itself.
+- Wire the aggregated non-blocking `reliability-gates` CI job (see Evidence Provenance And CI Wiring) so runtime and flake history accrue with machine provenance instead of hand-entered evidence.
+- Split the current integration branch into focused PRs before final review and merge. Do not land this as one large terminal reliability PR; each PR should have its own invariant, evidence command, residual gaps, dependency note, and follow-up relationship to the rest of the stack.
+- #7001 merged on 2026-07-03. Merge the gate-specific PRs only after resetting each from the worktree slice per the file assignment.
+- Before merging each child PR, recheck it against the latest #7001/main because the shared manifest file can create normal stack drift or conflicts even when the PR currently reports clean.
+- After merge, run the registered gate commands in soak and record runtime/flake evidence before promoting any gate to blocking.
+- Update the factory skills again after the repo-side gate commands and CI jobs are stable, so agents can invoke exact commands instead of only requiring reliability evidence.
+- Keep #7054's accepted gap visible: deterministic Claude-like TUI streams validated the renderer path, but live Claude Code was not part of that PR's validation. #7133 has since closed the alt-screen snapshot-restore hole in the same hidden-output machinery, and its buffer-merge root cause is exactly the class the `terminal-output.scrollback-restore` gate must make hard to reintroduce; the live-Claude validation gap remains.
+
+## PR Split Plan
+
+The current `brennanb2025/fix-terminal-reliability` branch is an integration branch, not the desired final review shape. It should be split into smaller PRs so reviewers can verify one reliability class at a time, bisect failures, and avoid merging a broad terminal lifecycle change without understanding which protection each file belongs to.
+
+Branch and manifest ownership rules:
+
+1. `brennanb2025/fix-terminal-reliability` is #7001's head branch. Only manifest, checker, checker-test, and plan-doc changes may be committed to it. Split items 2-6 go to their own branches, stacked on this branch and retargeted to `main` after #7001 merges.
+2. #7001 may only declare gates whose `testFiles` all exist on `origin/main`. The checker's declared-file rule makes this self-enforcing: a manifest entry whose test files land in a later PR cannot pass CI inside #7001.
+3. Every later PR owns its own manifest entries. It adds or upgrades the gates it implements — entry, `evidenceRuns`, and `assertionRefs` — in the same PR as the tests and hardening they reference. The manifest file is shared, but each entry has exactly one owning PR, and manifest conflicts are resolved by entry ownership, never by hand-merging divergent copies of an entry.
+4. Commandless registered-gap entries (`protection: "none"`, no test files) may ride in #7001 because they reference nothing that can drift.
+5. The integration worktree is the single authoritative copy of every overlapping file. Child branches that have diverged (for example, `resume-sleeping-agent-session.ts` on #7008's branch is ~156 diff lines behind this worktree) must be reset from the worktree slice, not reconciled by hand.
+
+Recommended split:
+
+1. Reliability gate manifest and plan docs.
+   This PR owns `config/reliability-gates.jsonc`, the manifest checker hardening, checker tests, and this plan document. It should merge first because later PRs reference gate ids, evidence metadata, and promotion rules from the manifest. It should not include terminal product behavior changes.
+
+2. Provider ownership and session replay safety.
+   This PR owns the agent/provider identity invariant: activation, restore, sleep, hibernate, dedupe, clearing, and reconnect paths must not replay a provider session that is already owned, queued, pending, retained, or live in the workspace. It should include the provider-session ownership tests, repeat-activation lower-layer proofs, automatic resume claim coverage, same-session/wrong-session hook coverage, and any narrowly required product hardening.
+
+3. Startup hydration, provider fanout, and targeted liveness.
+   This PR owns startup-adjacent PTY registry hydration, targeted `pty.hasPty(id)` liveness, no-hot `pty:listSessions()` paths, Resource Manager polling scoping, SSH/provider unknown-liveness semantics, degraded-daemon fail-closed behavior, and related tests. It should depend on the manifest PR and should explicitly say that live SSH, WSL, remote-runtime, and production startup reconcile remain follow-up gates unless implemented in the same PR.
+
+4. Terminal lifecycle observability.
+   This PR owns compact anomaly breadcrumbs and bounded lifecycle traces for terminal lifecycle diagnosis. It should make failures easier to debug without claiming to prevent them. It should prove that diagnostics are compact, deduped or bounded, privacy-conscious, and included in the right local crash/diagnostic surfaces. Diagnostics-bundle export and live forbidden-transition artifacts should remain visible follow-ups until proven.
+
+5. PTY output, IPC, and performance hardening.
+   This PR owns daemon stream batching, main-to-renderer pending output caps, ACK/backpressure behavior, replay queue bounds, runtime/mobile stream byte budgets, and related perf-sensitive contracts. It needs the hardest review because small behavior changes here can cause input latency, hidden-output, memory, or ordering regressions. It should include explicit runtime budgets and should not promote any perf gate to blocking without soak evidence.
+
+6. Live local PTY and E2E reliability slice.
+   This PR owns the first live Electron/local PTY liveness gate. It should stay small: spawn a real terminal, prove one active PTY, type through focused xterm, verify process-visible output, hide/restore the workspace without duplicating or losing the PTY, resize and verify applied size, then exit and verify cleanup. It should remain `experimental` until CI/runtime history proves it is not flaky. Tab switch, app restart, SSH, WSL, Windows ConPTY, and scrollback restore should be explicit follow-ups unless included with equally strong oracles.
+
+Mapping to the open stack and file assignment:
+
+The six split items and the five open child PRs overlap; this mapping is authoritative. Every working-tree file must land in exactly one PR below. Four files contain changes from multiple reliability classes and must be split by hunk, with the earlier PR in the merge order owning its hunks first: `config/reliability-gates.jsonc` (per-entry ownership), `src/main/ipc/pty.ts` and its test, `src/renderer/src/components/terminal-pane/pty-connection.ts` and its test, and `src/renderer/src/store/slices/terminals.ts`.
+
+1. #7001 (split item 1, manifest and plan docs): `config/scripts/check-reliability-gates.mjs`, `config/scripts/check-reliability-gates.test.mjs`, `docs/reference/reliability-gates-implementation-plan.md`, and the manifest restricted to gates whose test files exist on `origin/main` plus commandless gap entries.
+2. #7008 (split item 2, provider ownership and replay safety): `src/renderer/src/lib/resume-sleeping-agent-session.ts` and test, plus its ownership slice of `terminals.ts`. Reset the branch from the worktree slice; keep the PR number for review continuity.
+3. #7005 (snapshot freshness and targeted renderer liveness): `terminal-dead-session-reconcile.ts` and test, `use-terminal-pane-lifecycle.ts` and test, `use-terminal-pane-global-effects.test.ts`, and the `pty.hasPty` preload/API plumbing (`src/preload/api-types.ts`, `src/preload/index.ts`, `src/renderer/src/web/web-preload-api.ts`, the `hasPty` slice of `src/main/ipc/pty.ts`).
+4. New PR (split item 3, startup hydration, provider contracts, no-hot listing): `hydrate-local-pty-registry.ts` and test, `daemon-pty-adapter.ts` and test, `daemon-pty-router.ts` and test, `degraded-daemon-pty-provider.ts` and test, `pty-session-id.test.ts`, `src/shared/pty-session-id-format.ts`, `src/main/providers/ssh-pty-provider.ts`, `ResourceUsageStatusSegment.tsx`, `resource-manager-session-polling.test.ts` (currently untracked), `store-session-cascades.test.ts`, and the SSH/liveness slices of `pty.ts`, `pty-connection.ts`, and `terminals.ts`.
+5. New PR (split item 4, lifecycle observability): `terminal-lifecycle-diagnostics.ts` and `terminal-lifecycle-diagnostics.test.ts` (currently untracked).
+6. New PR (split item 5, PTY output/IPC/perf hardening): `daemon-stream-data-batcher.ts` and test, `terminal-output-batching.test.ts`, `terminal-subscribe-buffer.test.ts`, and the pending-output-cap and replay/ACK slices of `pty.ts` and `pty-connection.ts`.
+7. #7004 (xterm addon containment, outside the six split items): keep as its own PR; reset `pane-lifecycle.ts` and test from the worktree versions, and adopt fresh main's renamed WebGL test files during rebase.
+8. #7006 (visible geometry, outside the six split items): keep as its own PR; any geometry hunks in `pty-connection.ts` belong to it.
+9. #7007 (persisted upgrade fixtures): the gate is commandless with `protection: "none"`, so its manifest entry rides in #7001 under rule 4. Keep #7007 open only if it will deliver the actual fixture corpus and command; otherwise close it as absorbed.
+10. New PR (split item 6, live local PTY E2E slice): `tests/e2e/terminal-live-pty-liveness.spec.ts` (currently untracked).
+
+Merge order: #7001, then #7008, then #7005, then startup hydration/provider contracts, then lifecycle observability, then PTY output/IPC/perf hardening, then #7004, then #7006, then the live E2E slice. #7007 can merge any time after #7001 or be closed. The stack stays flat on #7001; each child rebases onto the merged state immediately before its own merge, and manifest conflicts are resolved by entry ownership.
+
+If one later PR reveals a missing invariant in an earlier PR, update the earlier manifest entry or PR description before promotion rather than silently broadening the later PR's claim. The split is part of the reliability plan because reviewability, dependency clarity, and targeted rollback are themselves reliability controls.
+
+Each split PR should follow Brennan's PR process before it is considered ready: write a scoped design doc, run delegated design review, implement from the reviewed doc, verify completeness against the original request and reliability gate, audit headline behavior, run `perf`, run the code-review/fix loop, run merge-confidence validation, open an unmerged PR, verify pushed state, and stabilize CodeRabbit/check findings. The PR text should describe that process as "Brennan's PR process"; do not mention the internal skill name in PR-facing text.
+
+## PR Proof Contract
+
+Every split PR must include a short proof section in its PR description before it can be treated as ready. This section should use direct evidence, not confidence language. Avoid absolute claims such as "this guarantees terminal reliability" or "this cannot regress performance." The valid claim is narrower: "this PR is safe to merge because the affected invariant, performance budget, and residual gaps were checked with the evidence below."
+
+Required PR description sections:
+
+- **Reliability invariant:** the class-level rule this PR protects and the recent issue, PR, or accepted gap that motivated it.
+- **Material impact:** why this will meaningfully reduce escaped regressions, including which known bug class becomes harder to reintroduce.
+- **Product change type:** docs/tooling, test-only gate, product hardening, diagnostics, perf hardening, or mixed.
+- **Performance risk inventory:** whether the PR touches typing, focus, tab/workspace switch, visibility resume, resize, render, startup, provider listing, hidden output, PTY output, store subscribers, git/worktree scans, SSH, WSL, Windows, or mobile/relay paths.
+- **No-regression evidence:** the exact count test, deterministic contract test, metric artifact, typecheck, or live run that proves the PR did not add unbounded scans, polling, hidden-pane wake loops, startup awaits, subprocess churn, renderer jank, or output buffering growth.
+- **Correctness evidence:** the exact oracle and command proving the invariant. Tests that only prove implementation shape are not enough.
+- **Provider/platform coverage:** local, daemon, SSH, WSL, remote-runtime, mobile/relay, macOS, Linux, and Windows must be marked covered, unaffected, or accepted-gap.
+- **Residual gaps:** what this PR still does not prove. These gaps must remain in the manifest until a later PR removes them.
+- **Promotion status:** whether the gate is `none`, `partial`, or `active`, and why it is not promoted beyond its current maturity.
+- **Rollback or demotion rule:** what signal should cause the PR's gate to be demoted, disabled, or followed up.
+
+Performance rule:
+
+- A reliability PR is incomplete if it adds polling, global session listing, provider fanout, hidden-pane renderer wakeups, startup-critical awaits, repeated resize IPC, subprocess churn, unbounded output queues, broad store subscriptions, or uncapped diagnostics without a deterministic count test or metric budget.
+- `pty:listSessions()` remains a global management/diagnostic operation, not a hot-path liveness primitive. Typing, focus, tab/workspace switch, visibility resume, resize, render, and per-pane liveness paths should use targeted APIs or cached ownership unless a PR provides measured proof that the broader call is safe.
+- Any PR touching git/worktree scans, SSH git providers, WSL paths, cleanup, or startup repository enumeration must include the `git-crash-perf` style proof: bounded subprocesses, timeout/abort behavior, provider parity, and no startup-critical scan.
+- Any PR touching terminal/session/provider/startup code must use the terminal-session reliability contract plus a performance budget. A green reliability test without a perf proof is not enough.
+- Any PR that changes PTY output, IPC batching, ACK/backpressure, replay, or hidden-output behavior (split item 5 especially) must include a before/after run of an existing terminal perf artifact — `test:e2e:terminal-perf:scale:report` or the typing-latency specs — on the same machine, with both results in the PR description. Deterministic byte/count contracts prove bounds; only the perf artifact proves interactive latency did not regress.
+- The gates themselves are a performance surface. Every command-backed gate must declare a runtime budget in the manifest, unit-layer gate commands should stay in single-digit seconds, and the aggregated CI job in Evidence Provenance And CI Wiring must stay inside its own wall-clock budget so reliability coverage does not degrade the development loop it protects.
+
+Per-PR proof expectations:
+
+1. Reliability gate manifest and plan docs.
+   Performance proof: no runtime product path changes. Checker runtime should stay cheap and deterministic, and manifest validation should not depend on network calls or live providers.
+   Materiality proof: the checker prevents overclaiming by requiring evidence metadata, existing test files, command-backed assertions, covered scope, and clear `none`/`partial`/`active` status.
+
+2. Provider ownership and session replay safety.
+   Performance proof: activation and replay checks must be bounded by indexed workspace-local state, not repeated provider listing or O(records x pending-startups) scans. Tests should include repeat activation and bounded-work assertions where scans are introduced.
+   Materiality proof: this blocks the class where an inactive or hidden but valid provider session is mistaken for unowned and resumed again.
+
+3. Startup hydration, provider fanout, and targeted liveness.
+   Performance proof: startup hydration must not block first usable UI, must expose bounded counters, and must not multiply provider/git scans across repos and worktrees without a budget. Hot interaction paths must prove zero broad `pty:listSessions()` calls.
+   Materiality proof: this prevents stale global snapshots or failed provider observations from closing or replacing the wrong terminal, while reducing session-listing fanout.
+
+4. Terminal lifecycle observability.
+   Performance proof: breadcrumbs and traces must be capped, primitive-only, deduped or ring-buffered, and must not serialize terminal output, filesystem trees, scrollback, or large state. Tests must prove caps and sanitization.
+   Materiality proof: this does not itself prevent regressions, but it materially shortens future diagnosis by preserving the lifecycle facts needed to debug ownership, liveness, restore, resize, and replay failures.
+
+5. PTY output, IPC, and performance hardening.
+   Performance proof: this PR must carry the strongest perf evidence in the stack: byte caps, queue caps, ACK/backpressure ordering, active-output priority, hidden-output behavior, and no unbounded renderer/main buffering. It should not become blocking until soak data shows stable runtime and no unexplained flakes.
+   Materiality proof: this protects the freeze/input-lag/memory-growth class where background or high-volume terminal output harms the active terminal or the app.
+
+6. Live local PTY and E2E reliability slice.
+   Performance proof: the Playwright gate must be small, serial where needed, artifact-producing, and timed. It should prove the user path without becoming a broad slow/flaky suite. Runtime and flake history are required before promotion.
+   Materiality proof: this covers the layer that fake providers cannot prove: real Electron terminal spawn, keyboard input, process-visible output, hide/restore, resize, and cleanup.
+
+## Factory And Skill Follow-Up
+
+The current factory skills are close enough that the next step should be targeted edits, not a large new skill stack. The terminal-specific skill already exists, and Brennan's PR process already requires design review, perf review, code review, validation, pushed-state checks, and PR stabilization. The missing factory behavior is that reviewers and validators should explicitly fail PRs that omit the PR Proof Contract above.
+
+Recommended skill/factory updates:
+
+1. Update `terminal-session-reliability`.
+   Add the PR Proof Contract as a required output for terminal/session/provider/startup PRs. The skill should require reliability invariant, material impact, performance risk inventory, no-regression evidence, provider/platform coverage, residual gaps, promotion status, and rollback/demotion rule.
+
+2. Update `review-code`.
+   Make missing proof-contract sections a review finding for P0-capable terminal/session/provider/startup/release surfaces. Reviewers should not accept "tests pass" as enough when the PR lacks a named invariant, perf budget, provider/platform matrix, or accepted-gap statement.
+
+3. Update `brennan-test-changes`.
+   Require validation reports to say whether the relevant reliability gate command ran, whether the evidence proves the user-visible invariant, whether the test layer is sufficient, and whether any skipped SSH, WSL, Windows, remote-runtime, mobile, or live-Electron path is an accepted gap.
+
+4. Update `playwright-reliability-tests`.
+   Add the live-terminal reliability bar directly: Playwright terminal tests must prove real input/output or visible degraded state, use deterministic event oracles instead of blind sleeps, record runtime/artifacts, and stay `experimental` until flake history supports promotion.
+
+5. Update `perf`.
+   Add the PR Proof Contract's no-regression evidence language so perf reviewers require deterministic count tests or metric artifacts for polling, global session listing, provider fanout, hidden-pane wakeups, startup awaits, subprocess churn, output queue growth, and broad store subscriptions.
+
+6. After the repo-side gate commands stabilize, update factory prompts to invoke exact gate ids and commands.
+   Do this after the PRs land or after their commands are stable enough to reference. Before then, the factory should require the gate class and evidence, but should not hard-code commands that may still split or rename during PR extraction.
+
+Do not add a separate broad "terminal reliability reviewer" skill unless the existing skills fail to enforce this contract after these edits. Too many overlapping skills can make agents diffuse responsibility; the better shape is one terminal reliability contract plus required calls from design, review, perf, Playwright, and validation stages.
+
+## Coverage Reality Check
+
+This first stack is useful, but it must not be described as "terminal reliability is covered." The executable protection in this branch covers a deliberately small first slice of recent escaped classes:
+
+- stale local/daemon liveness snapshots closing newer PTY bindings;
+- provider-session replay ownership for several active, inactive, queued, pending, live, retained, same-session hook, and wrong-session hook records, with real hook timing through Electron still explicitly partial;
+- targeted `hasPty` liveness after visibility resume instead of hot-path global `listSessions`;
+- degraded daemon fallback fail-closed behavior for restored worktree-scoped and legacy/non-scoped ids;
+- SSH provider listing/`hasPty` failures treated as unknown rather than destructive ownership evidence, plus mocked provider/renderer coverage for deferred SSH attach, passphrase cancellation, transient deferred reattach failure, and expired deferred relay fallback;
+- remote replay coalescing, FIFO ordering, and bounded burst-tail coalescing at the renderer unit layer;
+- visible geometry contract behavior for delayed split layout settle, 0x0 split-right recovery, applied-vs-requested `pty:getSize`, and resume-time size reassertion;
+- boot-time local PTY registry hydration counters for repo/worktree enumeration, skipped SSH repos, adapter/session listing, registration/skips, duration, and failure phase;
+- terminal lifecycle anomaly breadcrumbs recorded into renderer crash diagnostics with dedupe and compact identity fields;
+- live local Electron PTY spawn, active PTY listed exactly once, focused xterm keyboard input, process-visible output, repeated workspace hide/restore of the same PTY, actual resize propagation, and exit cleanup.
+
+Those gates should catch re-regressions in those classes, and the tests are intentionally deterministic. They mostly use renderer-unit, renderer-state, provider-contract, or fixture layers because those layers provide precise oracles with low flake risk. That is a strength for the classes they cover.
+
+The manifest also contains commandless registered gaps for persisted-session upgrade fixtures, Windows ConPTY, startup color-query handling, and other future gates. Those entries are useful planning structure, not executable protection. Visible geometry convergence, xterm addon containment, IME/native text forwarding, runtime/mobile streams, and startup hydration counters now have first executable contract slices, but they remain `experimental` until they have repeatable CI/runtime history and red-green evidence. Output backpressure now has daemon stream and main pending-output contract slices, but the broader live perf budget remains experimental until it has repeatable metric artifacts, runtime history, and red-green evidence. The live local PTY gate now has a first executable Playwright slice that includes repeated workspace hide/restore of the same PTY, but it remains `experimental` until it has repeatable CI/runtime history and red-green evidence.
+
+The critique that this stack is still shallow is fair. The current gates do not yet prove:
+
+- tab-switch restore, scrollback after restore, app restart persistence, and non-local live providers;
+- Windows ConPTY behavior, including shell resolution, cursor restore, rewrite repaint, CJK/IME edge cases, kitty/input protocol behavior, and activation hangs;
+- terminal input latency, throughput, hidden-output pressure, renderer CPU, and store/subscriber hot paths under load;
+- IME and synthetic input forwarding across macOS, Windows, Linux/fcitx/Sogou, Vietnamese, CJK, Arabic, paste, and modifier paths;
+- focus recovery after browser/terminal handoffs, tab switches, app restart, and non-local provider restore;
+- daemon degraded spawns, stale bootstrap, remote-runtime startup, SSH/WSL provider boundaries, and mirror polling;
+- scrollback clearing/restoration correctness across hidden output, snapshot replay, and restore.
+
+Those missing classes already appear in the full pain-points report, but this implementation plan should keep them visible so the first stack does not become false confidence.
+
+## Fresh-Eyes Corrections
+
+The follow-up reviews changed the plan in several concrete ways:
+
+- Windows/IME/CJK are not covered by the current executable stack. Orca has useful unit coverage and product logic for local Windows ConPTY compatibility, including withholding Kitty keyboard protocol for local native Windows panes, but no live Windows ConPTY gate proves post-agent ordinary keys, shell parity, CJK repaint, cursor restore, resize readback, or real IME composition.
+- SSH restore is better covered than the first manifest draft said, but only at store/provider/main/renderer contract layers. The current gate asserts deferred wake metadata, deferred attach/passphrase/expired-relay contracts, and unknown-liveness behavior with fake transports; it does not prove a live SSH relay, WSL, remote-runtime mirror polling, false-live visual state, or real network reconnect timing.
+- Perf/backpressure is currently deterministic byte-accounting proof, not live responsiveness proof. The branch asserts daemon `write(false)`/`drain` behavior, cross-session drain priority for flush-immediate output, pure-backpressure cleanup, and main pending-output caps; it does not yet prove renderer parse pressure, event-loop delay, active key latency, or live hidden-output flood behavior.
+- Daemon startup reconcile remains a mechanism, not a production-proven startup behavior. Either wire it with dry-run diagnostics and prior-worktree aliases, or keep it as an accepted gap with a named owner.
+- Gate evidence is still too easy to hand-enter. The checker now prevents obvious metadata overclaims, but it does not prove the registered command was run in CI, preserve artifacts, or verify that each declared assertion executed. Add machine-checkable artifact/CI-run provenance before promoting gates to blocking.
+- Manifest evidence in this worktree is not yet durable. Three referenced test files are untracked (`resource-manager-session-polling.test.ts`, `terminal-lifecycle-diagnostics.test.ts`, `tests/e2e/terminal-live-pty-liveness.spec.ts`) and roughly 39 more files carry uncommitted modifications. Until each slice is committed and pushed to its owning split branch, the manifest references proof that CI cannot run and that no one can reproduce from a ref.
+- Fresh `origin/main` renamed `terminal-webgl-paste-recovery.{ts,test.ts}` to `terminal-webgl-atlas-recovery.{ts,test.ts}` and added `pane-webgl-context-recovery.test.ts` and `pane-webgl-renderer.test.ts` in #6949. The `xterm-addon.boundary-containment` gate and its documented command reference the old filename and will fail the checker after rebase; the gate should adopt the renamed and new tests during the rebase. #7133 (merged 2026-07-02) further modified `pane-webgl-renderer.{ts,test.ts}` and added `pane-reveal-repaint.{ts,test.ts}` and `terminal-visibility-resume.{ts,test.ts}`; the containment gate should evaluate those reveal-hardening tests for adoption in the same sweep. This is a confirmed instance of the stale-branch risk, and it shows the fresh-main review must include a mechanical testFiles-vs-main sweep, not only behavioral spot checks.
+- This document itself drifted from the manifest it describes (a prior revision said 23 gates, 13 partial/10 none, while the manifest held 24 gates, 15 partial/9 none). Treat the manifest as authoritative and this document's counts as dated snapshots.
+- The Resource Manager status segment no longer polls broad `pty:listSessions()` for its closed-popover badge in this branch. Broad session inventory is now scoped to the open Resource Manager and explicit kill/restart refresh paths; the no-hot-listing plan still needs status/diagnostic budgets and broader Electron counters before promotion.
+- Boot-time local PTY registry hydration now exposes a debug snapshot and deterministic tests for retry state, SSH skip-before-worktree-enumeration, local registration counters, adapter/session listing counts, recoverable adapter-list failures, fatal hydration failures, router adapter fanout, duration, and large daemon lists. This is still only a startup-hydration counter slice; it does not prove focus, typing, tab switch, workspace switch, render, high-session provider fanout, or real Electron startup budgets.
+- Terminal lifecycle anomaly diagnostics now record compact crash breadcrumbs in addition to console warnings. This is useful support evidence, not yet the full pane lifecycle trace buffer requested in the pain-points report.
+
+## Fresh Main Architecture Findings
+
+These findings came from focused reviews of current terminal lifecycle, renderer/input/perf, and provider/platform/startup code against freshly fetched `origin/main`. They are not all regressions introduced by the reliability-gate branch. They are the places the plan must cover before claiming broad terminal reliability.
+
+Current coverage already exists in more places than the first gate stack showed:
+
+- `terminal-dead-session-reconcile.test.ts`, `resume-sleeping-agent-session.test.ts`, and `pty-connection.test.ts` have substantial deterministic lifecycle coverage. A focused run of those files passed 292 tests.
+- Existing E2E/perf tests include terminal typing latency, hidden TUI visual restore, long-table scroll restore, daemon slow-init PTY restore, SSH localhost, and terminal tab-switch SIGWINCH restore.
+- Existing scripts include terminal perf report commands such as `test:e2e:terminal-perf`, `test:e2e:terminal-perf:scale`, `test:e2e:terminal-perf:scale:report`, and `test:e2e:terminal-perf:check-report`.
+
+The problem is that these tests are not yet organized into reliability gates with named invariants, owners, maturity, runtime budgets, flake history, red/green evidence, and promotion rules. Some also need review before promotion because a useful regression test can still be too broad, too slow, or too flaky to be a blocking gate.
+
+Local validation evidence from this review pass:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/terminal-dead-session-reconcile.test.ts \
+  src/renderer/src/lib/resume-sleeping-agent-session.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts
+```
+
+Result on 2026-07-01: 3 test files passed, 292 tests passed, duration 6.04s. This asserts the current deterministic lifecycle/unit layer is healthy in this worktree; it does not prove the broader live/platform/provider/perf gaps.
+
+The provider-session ownership gate now includes the queued/pending resume bridge from fresh `origin/main`: automatic resumes thread `resumeProviderSession` through `pendingStartupByTabId`, claim `automaticAgentResumeClaimsByTabId`, and treat pending startup, time-bounded runtime claims, and live agent statuses as provider-session ownership. The queued/pending claim scan is indexed once per activation so replay protection does not rescan every pending startup for every sleeping record. It also asserts live hook evidence claims only the matching provider session; a wrong-session hook cannot claim a replayed sleeping record; and stale automatic bridge claims cannot suppress recovery forever.
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/lib/resume-sleeping-agent-session.test.ts
+```
+
+Result on 2026-07-01: 1 test file passed, 34 tests passed, duration 7.86s. This focused command asserts repeat activation or replayed sleeping records do not launch a second resume for the same provider session while the first resume is still pending, queued pending-startup provider-session ids are read once per activation rather than once per replayed record, a live hook for the same provider session owns the replay, and a wrong-session hook does not. It does not yet prove real hook timing through Electron or a full Electron workspace activation loop.
+
+Structural manifest validation also passed locally:
+
+```sh
+pnpm run check:reliability-gates
+```
+
+Result on 2026-07-01 after the first expansion: reliability gate manifest check passed for 22 gates. This validated the manifest schema/metadata shape only; it did not prove that the registered commands covered the full terminal runtime.
+
+After the broader architecture review and fresh-agent review, the manifest was expanded further, including a dedicated WSL restore contract instead of burying WSL under SSH/remote coverage, and later the three convergence/escape-detection gaps registered after the #7133/#7148 saga. As of 2026-07-03, main's manifest holds 27 gates split 9 `partial` (gates whose evidence exists on main itself, including the #7133/#7148/#7173/#7192 regression tests) and 18 `none`; the pending stack raises more gates to `partial` as its slices land. These counts are dated snapshots, and the manifest itself is authoritative. The new entries are intentionally `experimental`; a few now have first executable slices, while others are still commandless gap markers. They are not active protection until each gate gets an implemented command, red/green proof, runtime evidence, and promotion history.
+
+The manifest checker now enforces the review lessons that were easy to miss manually:
+
+- declared `testFiles` must exist;
+- every declared executable `testFiles` entry must be referenced by at least one gate command, so a manifest cannot borrow credibility from a file the registered command does not run;
+- command-backed gates cannot use title selectors such as `-t`, `--grep`, or `--testNamePattern`, because title drift silently changes coverage;
+- every gate must declare `protection: none | partial | active`;
+- commandless gates must declare `protection: "none"`, while command-backed gates must declare at least `protection: "partial"`;
+- `protection: "active"` is reserved for blocking gates with stable flake history and complete red/green evidence;
+- `partial` and `active` gates must include at least one passed structured `evidenceRuns` entry whose command exactly matches the gate command;
+- `none` gates must have no evidence runs, so planning gaps cannot look like tested coverage;
+- `partial` and `active` gates must include `assertionRefs` that point to declared test files and name the invariant assertions inside those files;
+- `none` gates must have no assertion refs, so planning gaps cannot borrow credibility from unrelated tests;
+- every gate must split risk scope from covered scope with `coveredPlatforms`, `coveredProviders`, and `coverageNotes`;
+- `coveredPlatforms` and `coveredProviders` must be inside the declared risk scope, and passed evidence-run platforms must appear in `coveredPlatforms`;
+- `soak` and `blocking` gates must have soaking/stable flake evidence and complete red/green evidence.
+
+How to read the manifest:
+
+- `platforms` and `providers` describe the affected risk scope for an invariant. They do not mean the current command exercises every listed platform/provider.
+- `coveredPlatforms`, `coveredProviders`, and `coverageNotes` describe the scope actually covered by current evidence. Empty `coveredProviders` on a partial gate usually means renderer/state/RPC contract coverage, not a real provider run.
+- `protection: "none"` means a registered gap only; `protection: "partial"` means useful executable coverage that is not blocking protection yet; `protection: "active"` means the gate is blocking and has stable evidence.
+- `evidenceRuns` are factual run records: date, runner, platform, exact command, result, duration, and summary. A local passed run supports `partial` coverage; it does not imply CI stability or flake history.
+- `assertionRefs` explain which assertion groups in the listed test files carry the gate. This is especially important for shared files like `pty-connection.test.ts`, where the whole file passing is not the same as the whole file being relevant.
+- `oracle`, `redGreenEvidence`, and `knownGaps` are the authoritative source for what is actually proved today.
+- Commandless `experimental` gates are registered gaps, not active protection.
+
+New or expanded manifest entries:
+
+- `terminal-platform.live-pty-liveness`
+- `terminal-platform.windows-conpty-liveness`
+- `terminal-performance.no-hot-list-sessions`
+- `terminal-performance.output-backpressure-budget`
+- `terminal-performance.input-throughput`
+- `terminal-performance.daemon-stream-backpressure`
+- `terminal-performance.store-and-git-hot-paths`
+- `terminal-provider.daemon-startup-degraded-contract`
+- `terminal-provider.ssh-remote-reattach-contract`
+- `xterm-addon.boundary-containment`
+- `terminal-output.scrollback-replay-fifo`
+- `terminal-output.scrollback-restore`
+- `terminal-input.ime-and-synthetic-forwarding`
+- `terminal-input.windows-conpty-keyboard-reset`
+- `terminal-render.windows-cjk-repaint`
+- `terminal-shell.windows-resolution-parity`
+- `terminal-capability.startup-color-query`
+- `terminal-runtime.mobile-stream-budget`
+- `terminal-mirror.parser-parity`
+- `terminal-observability.restore-convergence-selfcheck`
+- `terminal-render.pixel-refresh-repair`
+
+The visible geometry gate now has its first deterministic renderer/main contract slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/pty-size-reconcile.test.ts \
+  src/renderer/src/components/terminal-pane/split-right-white-screen.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts \
+  src/main/ipc/pty.test.ts
+```
+
+Result on 2026-07-02: 4 test files passed, 481 tests passed, duration 5.98s. This focused command asserts delayed hidden split-layout settles are forwarded instead of stopping on a fixed frame budget, unmeasurable frames do not count as settled, visible 0x0 split-right panes get a nonzero recovery size while hidden background 0x0 panes are not forced to desktop size, `pty:getSize` reports applied size when a provider can expose it, and visibility resume reasserts only real PTY/xterm drift without falling back to `listSessions`. It does not yet prove shell-visible `stty`, SSH/remote geometry, Windows ConPTY geometry/readback, or full Electron geometry under real split/tab/app-restart workflows.
+
+The live local PTY gate now has its first executable Electron slice:
+
+```sh
+pnpm run ensure:electron-runtime && npx playwright test \
+  tests/e2e/terminal-live-pty-liveness.spec.ts \
+  --config tests/playwright.config.ts --project electron-headless --workers=1
+```
+
+Result on 2026-07-02 after adding two workspace hide/restore cycles, exact active-PTY listing, and actual resize convergence: 1 Playwright test passed, test body 8.6s, full command 58.6s including build/setup. This one local macOS run asserts a real Electron local terminal binds a PTY, lists the active PTY id exactly once, accepts keyboard input through focused xterm, renders process output markers, preserves the same PTY id through repeated worktree switch-away/switch-back cycles, accepts additional keyboard input after each restore, applies an actual `pty:getSize` change after viewport resize, reports the same process-visible size, exits the probe, exits the shell, and removes the PTY id from `pty:listSessions`. It does not yet prove tab switch within the same worktree, scrollback after restore, app restart persistence, daemon, SSH, WSL, remote-runtime, Linux CI stability, or Windows ConPTY behavior.
+
+The IME/native-text gate now has its first deterministic renderer-unit slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/terminal-ime-native-text-forwarder.test.ts \
+  src/renderer/src/components/terminal-pane/terminal-ime-input-source.test.ts \
+  src/renderer/src/components/terminal-pane/terminal-paste-runtime.test.ts
+```
+
+Result on 2026-07-02: 3 test files passed, 63 tests passed, duration 887ms. This registers existing native-text, input-source, and paste/runtime forwarding contracts as a reliability gate slice. It does not prove real OS IME automation, Windows ConPTY post-agent keyboard reset, Arabic/RTL, JIS-yen, or the full CJK/Vietnamese/platform matrix.
+
+The xterm addon containment gate now has its first deterministic renderer-unit slice and product hardening for core addon load failures:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/lib/pane-manager/pane-lifecycle.test.ts \
+  src/renderer/src/lib/pane-manager/terminal-link-provider-guard.test.ts \
+  src/renderer/src/components/terminal-search-safe-find.test.ts \
+  src/renderer/src/lib/pane-manager/pane-webgl-refresh-lifecycle.test.ts \
+  src/renderer/src/components/terminal-pane/terminal-webgl-paste-recovery.test.ts
+```
+
+Result on 2026-07-02: 5 test files passed, 39 tests passed, duration 370ms. This focused command asserts a core addon `loadAddon` throw is pane-scoped and later addons still load, link provider throws degrade to no-link results, search decoration positive-integer crashes return `false` instead of escaping, and WebGL attach/refresh/recovery failures stay contained. It does not yet prove typed PTY input/output survives a live addon failure in Electron, nor WebGL dispose/reset throws across active, hidden, and resumed panes.
+
+The runtime/mobile stream budget gate now has its first deterministic runtime-RPC slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/runtime/rpc/terminal-subscribe-buffer.test.ts \
+  src/main/runtime/rpc/terminal-output-batching.test.ts \
+  src/main/runtime/rpc/terminal-multiplex.test.ts
+```
+
+Result on 2026-07-02: 3 test files passed, 29 tests passed, duration 2.78s. The covered contracts assert mobile initial snapshots downgrade until they fit the 512KB budget, requested binary snapshots downgrade until they fit the 2MB budget, binary live output queued while the initial snapshot is serializing stays within 256KB while preserving the newest tail, large binary output is split into 48KB-or-smaller frames, output bursts are coalesced before emission, aborted subscribes do not register stale listeners, and stale mobile resize re-stream completions are dropped. It does not yet decide JSON subscribe fallback parity/deprecation.
+
+The replay/scrollback gate now has its first executable slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts
+```
+
+Result on 2026-07-02: 1 test file passed as part of the focused pty-connection validation, with 255 tests in that file. The relevant replay assertions cover four replay invariants: remote replay payloads that overlap before parsing starts intentionally coalesce to the latest payload, a replay whose xterm parsing has already started is not overwritten by a newer replay, multiple replay notifications accepted after parsing starts drain FIFO, and burst tails are bounded while keeping the newest snapshot. It does not yet prove fresh-main `clearBeforeReplay` metadata, metadata-only replay, or hidden-output/live-output interleaving.
+
+The no-hot-`listSessions` gate now has its first executable slice and product hardening. The renderer/preload/main path exposes `pty.hasPty(id)` and uses targeted liveness on visibility resume and first input after resume, instead of calling global `pty:listSessions()` from that hot path. The resize-on-resume slice also asserts it uses targeted `getSize` plus `resize` rather than broad session listing. Light tab switches and visible active-state resume now assert they do not call `pty:listSessions`, `pty.hasPty`, or `pty.getSize`; a representative active-PTY case proves the scheduler hint `setActiveRendererPty` is the only allowed PTY API call there. Resource Manager broad session inventory is now scoped to the open popover instead of polling for its closed badge.
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/terminal-dead-session-reconcile.test.ts \
+  src/renderer/src/components/terminal-pane/use-terminal-pane-lifecycle.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts \
+  src/renderer/src/components/status-bar/resource-manager-session-polling.test.ts \
+  src/renderer/src/components/terminal-pane/use-terminal-pane-global-effects.test.ts
+```
+
+Result on 2026-07-02: 5 test files passed, 321 tests passed, duration 6.34s. The focused no-hot command asserts visibility resume prefers targeted `hasPty` even when `listSessions` is available, first input after visibility resume calls targeted `hasPty` once, resize re-assertion after visibility resume uses `getSize`/`resize` without `listSessions`, light tab switches and visible active-state resume avoid `listSessions`/`hasPty`/`getSize` fanout while still allowing the active PTY scheduler hint, SSH/remote broad listing is skipped, Resource Manager broad session inventory polling is scoped to the open popover, panes close only on authoritative `false`, and unknown liveness keeps panes alive. It does not yet count raw focus, workspace switch, render, high-session typing, or real provider fanout in Electron.
+
+The store/git hot-path gate now has its first boot-hydration counter slice. This is not broad interactive perf coverage; it is the narrow startup-adjacent part that previously had no measurable budget.
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/memory/hydrate-local-pty-registry.test.ts
+```
+
+Result on 2026-07-02: 1 test file passed, 8 tests passed, duration 0.342s. The focused command asserts provider-unavailable hydration does not scan repos/worktrees/providers and remains retryable, SSH repos are skipped before worktree enumeration, local daemon sessions are registered with debug counters for repo/worktree/adapter/session/register/skip/duration fields, recoverable adapter-list failures are counted without failing the whole hydration pass, fatal hydration failures record failed phase and error message, router-backed hydration records multi-adapter fanout counts, pre-existing pid rows are not clobbered, and large daemon session lists hydrate without spreading into argument-limit failures. It does not yet count store recomputes, git status requests, provider listings, focus, typing, tab switch, workspace switch, render, or high-session Electron fanout.
+
+The degraded-daemon provider contract now has its first executable slice and product hardening. Existing-session operations fail closed when ownership is unknown, and restored daemon session ids, including legacy/non-worktree-scoped ids, cannot silently spawn a local fallback PTY under the old id. Fresh degraded-mode PTY creation still intentionally routes through fallback when the caller marks it as new, and process inspection returns benign defaults for unknown ownership instead of leaking fail-closed routing errors into window-close flows.
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/daemon/degraded-daemon-pty-provider.test.ts \
+  src/main/daemon/daemon-pty-adapter.test.ts \
+  src/main/daemon/daemon-pty-router.test.ts
+```
+
+Result on 2026-07-02: 3 test files passed, 107 tests passed, duration 2.54s. Focused provider validation asserts discovered daemon sessions route to the daemon, fresh degraded-mode PTYs route to fallback only when marked new, router spawn discovers uncached existing sessions before choosing an adapter, router spawn fails closed instead of falling through to current when legacy ownership cannot be listed or a known session has exited, targeted `hasPty` discovery caches legacy ownership before later operations, real daemon adapter `listProcesses()` discovery seeds targeted `hasPty` liveness, folder and floating terminal workspace ids survive startup reconcile when valid, restored worktree-scoped and legacy/non-scoped ids do not fall back through spawn after ownership is lost or unknown, unknown or exited existing ids fail closed for routing operations, startup reconcile can dry-run orphan detection without killing live daemon sessions, and process inspection returns benign defaults for unknown ownership. It does not yet prove production startup reconcile wiring, prior-worktree aliases, or real daemon process restart behavior.
+
+The SSH/remote provider contract now has its first executable provider/main/renderer slice. Provider observation failures are treated as unknown instead of destructive evidence: a failed SSH `listProcesses()` call does not clear previously learned ownership, and a rejected SSH `hasPty()` probe returns unknown liveness (`null`) rather than dead. The focused renderer/provider tests also assert that saved SSH sessions reattach through relay `pty.attach`, expired relay attach does not silently fresh-spawn inside the provider, deferred passphrase cancellation does not auto-reconnect, saved leaf/tab session ids are used after connection, transient deferred reattach failure preserves the saved session id without clearing pane/tab bindings or fresh-spawning, deferred SSH no-result cleanup clears the pending serializer without consuming the saved restore id, and expired deferred relay state clears stale pane/tab bindings before one fresh replacement spawn.
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/providers/ssh-pty-provider.test.ts \
+  src/main/ipc/pty.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts \
+  src/renderer/src/store/slices/store-session-cascades.test.ts
+```
+
+Result on 2026-07-02: 4 test files passed, 546 tests passed, duration 6.48s. Focused validation asserts store wake-hint metadata for disconnected SSH relay ids, provider attach/expired-attach behavior, main-process ownership and targeted-liveness failure semantics including async provider rejection, mocked-renderer deferred SSH attach/transient-failure/expired-relay fallback, and deferred SSH no-result cleanup that clears the pending serializer without consuming the saved restore id. It does not yet prove WSL, remote-runtime mirror polling, a live SSH relay, pending-vs-attached UI status, or real network reconnect timing.
+
+API-surface validation after adding `pty.hasPty(id)`:
+
+```sh
+pnpm run typecheck:node
+pnpm run typecheck:web
+```
+
+Result on 2026-07-02: both typechecks passed.
+
+The output-backpressure budget gate now has a deterministic provider/IPC/renderer slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/daemon/daemon-stream-data-batcher.test.ts \
+  src/main/ipc/pty.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts
+```
+
+Result on 2026-07-02: 3 test files passed, 469 tests passed, duration 9.35s. The focused command asserts daemon socket `write(false)`/`drain` ordering, cross-session interactive output priority, pure-backpressure cleanup, bounded daemon queued bytes, main pending renderer caps, active pending-output protection, replay/backlog slices, and ACK-gated in-flight bounds. It does not yet prove live hidden-output floods, renderer parse pressure, event-loop delay, active key latency, or full Electron perf artifacts.
+
+Combined focused terminal/provider validation:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/main/daemon/degraded-daemon-pty-provider.test.ts \
+  src/main/ipc/pty.test.ts \
+  src/renderer/src/components/terminal-pane/terminal-dead-session-reconcile.test.ts \
+  src/renderer/src/components/terminal-pane/use-terminal-pane-lifecycle.test.ts \
+  src/renderer/src/components/terminal-pane/pty-connection.test.ts \
+  src/renderer/src/lib/resume-sleeping-agent-session.test.ts
+```
+
+Result on 2026-07-01: 8 test files passed, 625 tests passed, duration 6.55s. This was focused terminal/provider validation, not structural manifest validation.
+
+Reliability manifest checker validation is separate:
+
+```sh
+pnpm run check:reliability-gates
+```
+
+Result on 2026-07-02: reliability gate manifest check passed for 27 gates. This validates the manifest shape and referenced files; it does not prove the registered commands cover the full terminal runtime.
+
+The highest-risk architecture boundaries are:
+
+- Terminal/session ownership is spread across tab PTY ids, leaf PTY ids, last-known relay ids, pending reconnect maps, deferred SSH maps, sleeping records, and provider ownership. The plan needs a derived ownership snapshot oracle for tests/logging so attach, reattach, exit, clear, hibernate, and restore can prove all authorities agree.
+- Queued or pending provider sessions must count as owned. This branch now restores the pending startup and automatic resume claim bridge and covers same-session/wrong-session hook evidence, but real hook timing and repeat activation through Electron still need coverage. This is the class-level invariant motivated by #6800.
+- Stale branch risk is real. The reliability branch previously lacked some protections present on fresh `origin/main`, including queued automatic agent resume claims and exact `hasPty` missing-session reconciliation. This branch has restored those two deterministic slices; the full stack still needs final rebase/retarget verification before landing.
+- Main PTY ownership historically had convenient fallback behavior for unknown ownership. This branch now hardens daemon-router and degraded-provider paths so discovered legacy ownership is cached and unknown existing-session operations fail closed, but SSH/remote/diagnostic listing paths still need live coverage.
+- Startup daemon reconciliation is present as a mechanism but not wired as production startup behavior. Without it, the daemon can still have a session while the adapter no longer knows it owns that session.
+- Degraded daemon routing on fresh `origin/main` can fall back to local for unknown existing-looking sessions. This branch now hardens operation plus restored worktree-scoped and legacy/non-scoped spawn paths to fail closed and registers a focused provider contract, but startup reconcile, prior-worktree aliases, and real daemon restart behavior still need coverage.
+- `pty:listSessions` both observes and mutates provider ownership. This branch now asserts failed SSH observation does not clear previously learned ownership and mocked deferred SSH attach/transient-failure/expired-relay fallback behaves correctly, but WSL, remote-runtime mirror polling, live SSH reconnect timing, and false-live tab behavior remain unproved.
+- Resource Manager broad session inventory is now limited to popover-open and explicit management refreshes. It still needs a status/diagnostic budget proving open-popover polling is bounded with many providers and does not become a hot-path substitute.
+- Boot hydration (`hydrateLocalPtyRegistryAtBoot`) is fire-and-forget, and this branch now records startup phase counters for repo/worktree enumeration, skipped remote repos, adapter/session listing, registrations/skips, duration, and failures. It can still create startup-adjacent git subprocess and daemon-listing churn across many local repos, so do not treat startup reliability as covered until those counters are budgeted in CI/runtime evidence.
+- Deferred SSH restore intentionally marks a saved PTY id before actual attach. This branch now covers the mocked renderer contract for passphrase cancellation, attach after connection, transient attach failure preserving retry state, and expired relay fallback; remaining work is false-live visual state, live SSH relay timing, and duplicate-spawn prevention under real reconnect churn.
+- Geometry truth is incomplete across SSH/remote paths. Requested size, renderer size, provider-applied size, shell-visible size, and the runtime mirror's parse dimensions must be separated instead of treated as interchangeable. #7192 proved the mirror can silently stay at birth dimensions while every other authority resizes, corrupting every later restore snapshot, and that mirror reflow must stay ordered with queued output writes.
+- Production diagnostics are still incomplete. This branch now records compact crash breadcrumbs for explicit terminal lifecycle anomalies, but real user reports still need a bounded pane lifecycle trace for ordinary transitions and a diagnostics-bundle artifact proving those breadcrumbs/traces are included.
+
+The lifecycle breadcrumb gate now has its first executable slice:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/terminal-lifecycle-diagnostics.test.ts
+```
+
+Result on 2026-07-02: 1 test file passed, 2 tests passed, duration 0.593s. The focused command asserts `warnTerminalLifecycleAnomaly` preserves its console warning, records a compact `terminal_lifecycle_anomaly` renderer crash breadcrumb with terminal identity/provider/PTY/reason fields, and dedupes repeated anomalies by lifecycle identity. It does not yet prove a full pane lifecycle trace buffer, diagnostics-bundle export, or forbidden-transition assertions across live Electron/provider flows.
+
+Specific current-source evidence checked in this pass:
+
+- Fresh `origin/main` has queued provider-session protections in `src/renderer/src/lib/resume-sleeping-agent-session.ts`: `resumeProviderSession`, `claimAutomaticAgentResume`, and `automaticAgentResumeClaimsByTabId`.
+- Fresh `origin/main` has exact liveness reconciliation in `src/renderer/src/components/terminal-pane/terminal-dead-session-reconcile.ts` and the `pty:hasPty` IPC handler in `src/main/ipc/pty.ts`.
+- Fresh `origin/main` still uses a single `pendingReplayData` slot in `src/renderer/src/components/terminal-pane/pty-connection.ts` (re-verified after #7133 merged; #7133 changed the same file's `applyMainBufferSnapshot` alt-screen restore path, not replay draining). This branch changes replay draining to coalesce snapshots before xterm parsing starts, queue accepted snapshots FIFO once parsing is in progress, and coalesce burst tails to the newest snapshot once the in-flight queue is full. The broader scrollback gate still needs metadata-only replay, `clearBeforeReplay`, and hidden/live-output interleaving coverage.
+- Fresh `origin/main` still has `DaemonPtyAdapter.reconcileOnStartup` with the explicit note that it has no production caller yet. This branch adds a dry-run mode so future startup wiring can audit would-kill sessions before destructive cleanup, but production startup-persistence remains a real gap until wired with prior-worktree aliases or tracked as an accepted gap with a gate.
+- Fresh `origin/main` still lets `DegradedDaemonPtyProvider.providerFor()` fall back to local when an existing-looking session id is unmapped and not rediscovered by `hasPty`. This branch changes existing-session operation paths and restored worktree-scoped plus legacy/non-scoped spawn paths to fail closed, adds contract tests, and makes real daemon adapter `listProcesses()` discovery seed targeted `hasPty` liveness. Startup reconcile wiring and live daemon-restart validation are still required before the full provider lifecycle is covered.
+- Fresh `origin/main` let the daemon router spawn an unknown existing `sessionId` on the current daemon when routing cache/discovery missed its real owner. This branch now probes/discovers all adapters first, requires `isNewSession` for intentional replacement after an authoritative exit event, and fails closed if legacy ownership cannot be listed or the known id has exited, so a stale existing id cannot silently mint over another daemon's history on current.
+- Fresh `origin/main` let canonical daemon-worktree restores treat a non-minted stale/local PTY id as restorable metadata. This branch now rejects non-minted restored ids when the workspace owner is canonical (`repo::path`, `folder:<id>`, or `global-floating-terminal`), clears the stale pane/tab binding, and fresh-spawns instead of calling daemon reattach or attaching that stale id. Shared parsing also recognizes folder and floating minted ids so startup reconcile does not falsely reap valid non-repo terminal sessions.
+- Fresh `origin/main` still lets `pty:listSessions` rebuild provider ownership while SSH listing failures become empty lists. This branch adds store/main/provider/renderer contracts proving disconnected SSH relay ids are retained as deferred reconnect metadata and sidebar wake hints rather than attached PTY proof, failed SSH observation does not clear previously learned ownership, `hasPty` failure is unknown, saved SSH sessions attach through relay `pty.attach`, deferred passphrase cancellation does not auto-reconnect, transient deferred reattach failure preserves the saved session id, deferred SSH no-result cleanup clears pending serializer state without consuming retry metadata, and expired deferred relay state clears stale pane/tab bindings before one fresh replacement spawn. Live SSH, WSL, remote-runtime mirror polling, false-live visual state, and real reconnect timing remain unproved.
+- Fresh `origin/main` has strong Windows/unit coverage around ConPTY packaging, shell resolution, size validation, keyboard reset, complex-script classification, and Windows PTY compatibility, but the real Windows Electron/ConPTY path is still not covered as a reliable PR gate.
+- Fresh `origin/main` still exposes global `pty:listSessions`; this branch now also uses targeted `hasPty` for visibility/input liveness, but gates should keep typing, focus, tab switch, visibility resume, resize, and render paths from reintroducing broad daemon/provider listing.
+- Fresh `origin/main` has main-to-renderer ACK/backpressure, renderer output-scheduler caps, terminal perf scripts, and a daily/manual terminal perf workflow. This branch adds daemon stream batcher assertions that socket `write(false)` pauses further stream events until `drain`, flush-immediate output from another session drains before unrelated background backlog, global cleanup clears clients that only have backpressured writes, and queued daemon stream bytes stay bounded by preserving priority output plus the newest tail under sustained pressure; it also caps main pending renderer output per PTY and in total, trims background pending output before active pending output under total pressure, and asserts `seq`/`rawLength` metadata stays correct for the surviving tail. Live hidden-output pressure, active input latency, and full Electron perf artifacts remain unproved.
+- Fresh `origin/main` has binary runtime/mobile terminal stream caps, but the legacy JSON subscription path has weaker buffering/backpressure guarantees. Either gate both paths or explicitly deprecate/accept the JSON gap.
+
+Existing E2E/perf tests that should be reviewed for promotion:
+
+- `tests/e2e/daemon-slow-init-pty-gate.spec.ts`: strong daemon live-session restore signal; already has a tight user-visible invariant.
+- `tests/e2e/daemon-live-session-preservation.spec.ts` and `tests/e2e/daemon-slow-health-check-preservation.spec.ts`: likely useful daemon liveness/preservation gates after oracle and flake review.
+- `tests/e2e/terminal-typing-latency.spec.ts`, `tests/e2e/terminal-codex-local-typing-latency.spec.ts`, and `tests/e2e/terminal-history-size-typing-latency.spec.ts`: useful perf signals, but should be tied to stable budgets and artifacts before blocking promotion.
+- `tests/e2e/terminal-output-scheduler.spec.ts`, `tests/e2e/artificial-opencode-terminal-load.spec.ts`, and hidden-pressure scenarios: useful for output/backpressure, but should become metric-gates rather than visual-only broad E2Es.
+- `tests/e2e/terminal-hidden-tui-visual-restore.spec.ts`, `tests/e2e/terminal-document-visibility-webgl-recovery.spec.ts`, and `tests/e2e/terminal-tab-switch-visual-restore.spec.ts`: useful rendering recovery coverage; must keep deterministic text/buffer oracles and limit screenshots to diagnostics.
+- `tests/e2e/terminal-long-table-scroll-restore.spec.ts` and `tests/e2e/terminal-raw-emoji-table-scroll-restore.spec.ts`: useful scrollback/rendering regression evidence, but they contain timed waits and should be reviewed before blocking promotion.
+- `tests/e2e/ssh-localhost.spec.ts`, `tests/e2e/ssh-docker-relay-perf.spec.ts`, and SSH Codex replay/reconnect specs: useful SSH coverage, but should be separated into deterministic provider contracts plus a small number of environment-dependent live SSH soaks.
+- `tests/e2e/terminal-windows-shell-paste-ownership.spec.ts` and Windows env/icon tests: useful Windows path coverage, but not a substitute for a Windows ConPTY live liveness gate.
+
+Promotion split for existing E2Es:
+
+- Promote after soak: a smaller derivative of `terminal-typing-latency.spec.ts` as the live local PTY input/output gate.
+- Promote after soak: daemon preservation specs, especially slow-init and stale launch identity, as daemon warm-reattach gates.
+- Promote targeted: paste ownership tests where the oracle is visible output plus exactly-one PTY write.
+- Keep in soak/nightly: artificial OpenCode load, terminal scale perf report, hidden real-PTY pressure, and SSH Docker perf.
+- Keep as release evidence: raw emoji/golden rendering, long-table rendering, WebGL recovery, and Codex/OpenCode artifact repros.
+- Rewrite before blocking: column desync specs need resize-applied acknowledgement or an in-PTY `SIGWINCH` marker instead of post-resize sleeps.
+- Rewrite before blocking: SSH Codex artifact repros and headful dead-terminal repros need deterministic fixtures/events instead of long waits and screenshots as primary proof.
+
+The first live PTY gate is deliberately small:
+
+- File: `tests/e2e/terminal-live-pty-liveness.spec.ts`.
+- Command: `pnpm run ensure:electron-runtime && npx playwright test tests/e2e/terminal-live-pty-liveness.spec.ts --config tests/playwright.config.ts --project electron-headless --workers=1`.
+- Invariant: a newly opened local terminal has one active PTY id listed exactly once, accepts real keyboard input through focused xterm, delivers process output visibly, keeps the same PTY id through repeated workspace hide/restore cycles, accepts input after restore, applies an actual size change, and exits without leaving a stale binding.
+- Oracle: start a deterministic Node raw-mode probe, wait for `LIVE_PTY_READY_<runId>`, require the active PTY id to appear exactly once in `pty:listSessions`, type through `.xterm-helper-textarea`, require ordered per-key markers in serialized terminal content, switch to another worktree and back twice, require the active pane to keep the original PTY id, type again after each restore, resize and require `pty:getSize` to change plus a matching process-visible size marker, then exit the probe and shell and require the old PTY id to disappear from `pty:listSessions`.
+- Wait rule: no blind sleeps. Wait only on PTY binding, active worktree changes, ready marker, key markers, resize marker, and list-session absence.
+- Budget: target p95 under 75s on Linux including e2e build/setup, hard test timeout 90s. Start blocking on Linux only after soak; run macOS/Windows nightly until each platform has its own history.
+- Failure artifacts: PTY id, provider classification, pid if available, size snapshots, key latencies, final `listSessions`, trace, and screenshot.
+- Red/green proof: intentionally break xterm focus, PTY write forwarding, resize forwarding, and exit cleanup in separate local runs and confirm the gate fails for each class.
+- Local run on 2026-07-02: 1 Playwright test passed in 58.6s total, with an 8.6s test body, on this macOS worktree. This is one green local run for the local live PTY path, not enough flake/runtime history for blocking promotion.
+
+## Gate Coverage Matrix
+
+This is the intended shape after rebasing the stack. The first two columns are what we have now; the last two are what must be added or promoted before the plan earns broader reliability claims.
+
+| Reliability class | Existing evidence | Gap | Next gate or action |
+| --- | --- | --- | --- |
+| Stale liveness snapshots closing newer PTYs | `terminal-session.snapshot-freshness`; deterministic unit coverage | No full Electron tab-survival/input echo proof | Add `terminal-platform.live-pty-liveness` and keep snapshot gate as lower-layer guard |
+| Provider-session replay ownership | `agent-session.provider-ownership`; renderer state tests; queued/pending resume claim proof; bounded queued-claim index proof; same-session and wrong-session hook proof | Needs real Electron repeated activation/hook timing and bounded hook/status scan assertions if those grow | Add Electron activation proof |
+| Exact PTY liveness | Fresh main has targeted `hasPty` path | Current branch must not lose it; provider failures must be unknown, not dead | Rebase/restore exact liveness reconciliation and register it as a gate |
+| Store/session ownership coherence | Many focused tests touch pieces | No single derived oracle across tab, leaf, relay, reconnect, deferred SSH, and sleeping maps | Add ownership snapshot helper for tests/logging |
+| Terminal lifecycle diagnostics | `terminal-observability.lifecycle-breadcrumbs` now records compact anomaly breadcrumbs into crash diagnostics | No full pane lifecycle trace buffer or diagnostics-bundle artifact proof | Add bounded trace buffer and bundle artifact test |
+| Xterm addon/search/WebGL containment | `xterm-addon.boundary-containment`; deterministic core addon load, link-provider, search, and WebGL containment tests; #7054 hidden TUI product hardening | Needs live Electron typed input/output survival after addon failure, and WebGL dispose/reset throw coverage across active/hidden/resumed panes | Add focused component/live input echo follow-up; promote hidden TUI E2E after flake review |
+| Visible terminal geometry | #7006 renderer/provider-style gate; #7192's merged mirror-resize and reflow-ordering red tests on main | SSH/remote and shell-visible truth not fully covered; mirror-geometry slice unregistered until rebase | Add live PTY size readback and provider-applied-size contract; register #7192's mirror-authority tests |
+| Persisted session upgrade | #7007 fixture gate | Needs broader old-version corpus and daemon/SSH cases | Build immutable fixture corpus with startup timing |
+| Live local PTY lifecycle | `terminal-platform.live-pty-liveness` now has a focused Electron Playwright slice covering local spawn, single active PTY listing, xterm input, PTY output, repeated workspace hide/restore, actual resize propagation, and cleanup | Needs flake/runtime history, red-green evidence, tab switch, scrollback after restore, app restart persistence, and non-local provider coverage | Promote first slice after soak; add tab-switch/scrollback/restart follow-ups |
+| Windows ConPTY lifecycle | Recent fixes on main, some unit/platform logic | No required Windows live gate for ConPTY echo/resize/cursor/CJK/IME | Add `terminal-platform.windows-conpty-liveness` |
+| Windows keyboard reset | Unit coverage and recent #6999/#6858 fixes | No live proof that Enter/Backspace/Arrow return to normal after agent/TUI exit | Add `terminal-input.windows-conpty-keyboard-reset` |
+| Windows CJK/repaint | Complex-script and compatibility tests plus recent CJK repaint fixes | No Windows visual/pixel oracle for wide-glyph in-place redraws | Add `terminal-render.windows-cjk-repaint` |
+| Windows shell parity | Shell resolution unit coverage and packaging fixes | No local-vs-daemon parity gate for PowerShell 5/7, cmd, Git Bash, WSL, startup command delivery | Add `terminal-shell.windows-resolution-parity` |
+| Input latency/perf | Existing terminal perf scripts and typing-latency spec | Not registered, not tied to budgets/owners/promotion | Add `terminal-performance.input-throughput` |
+| Hot interaction listing | Recent #7002 fix, targeted `hasPty` path, deterministic visibility/input/resize no-listing assertions, light tab/active-state resume no-provider-fanout assertions, and closed Resource Manager broad-listing avoidance | No count gate proving `pty:listSessions` stays out of raw focus/workspace-switch/render/high-session Electron scenarios; open Resource Manager polling still needs an explicit budget | Extend `terminal-performance.no-hot-list-sessions` |
+| Daemon stream backpressure | Batching/chunking plus new batcher contract proof for `write(false)`/`drain`, cross-session flush-immediate drain priority, pure-backpressure cleanup, and bounded queued tails | No live hidden-output flood proof or active input latency artifact | Extend `terminal-performance.output-backpressure-budget` |
+| Renderer/main ACK and backlog | Main ACK gates, renderer scheduler caps, focused unit coverage, and bounded main pending tails exist | No live metric artifact tying hidden restore, active key latency, queued chars, and dropped backlogs to promotion | Extend `terminal-performance.output-backpressure-budget` |
+| Runtime/mobile terminal stream | `terminal-runtime.mobile-stream-budget`; runtime-RPC tests assert mobile initial snapshot <=512KB, requested binary snapshot <=2MB, pending live output <=256KB while snapshot loads, output frames <=48KB, output coalescing, abort cleanup, and stale resize re-stream suppression | Legacy JSON path parity/deprecation is undecided | Decide/gate/deprecate JSON fallback |
+| Store/git hot paths around terminal changes | Perf/git-crash skills exist; git status limits/coalescing exist; `terminal-performance.store-and-git-hot-paths` now has a boot-hydration counter slice | No Electron count gate for terminal-adjacent store projection, git status, focus, typing, tab/workspace switch, render, or high-session provider fanout | Extend `terminal-performance.store-and-git-hot-paths` with interactive counters |
+| IME and synthetic input | Input forwarding code and scattered coverage | No representative matrix or byte/cell oracle | Add `terminal-input.ime-and-synthetic-forwarding` |
+| Daemon degraded/startup provider contract | Mechanisms exist; daemon slow-init E2E helps | Startup reconcile is unwired; degraded fallback needs fail-closed/probe proof | Add daemon degraded and startup reconcile contract gates |
+| SSH/remote restore and mirror polling | SSH localhost and remote fixes exist; this branch has store/provider/main/renderer proof for deferred SSH wake metadata, SSH attach, listing failure, deferred attach, passphrase cancellation, transient deferred reattach failure, and expired deferred relay fallback | Remote mirror polling, live SSH relay behavior, false-live visual state, pending-vs-attached UI status, and real reconnect timing remain unproved | Add live SSH soak and remote mirror polling gates |
+| WSL restore and launch identity | `terminal-provider.wsl-restore-contract` is now registered as a first-class gap | No WSL provider contract or live Windows WSL smoke yet proves cwd/path identity, startup command delivery, targeted liveness, or restore ownership | Add deterministic WSL provider contract, then one focused Windows WSL smoke |
+| Scrollback/replay restore | Recent #7012 fix, long-table restore coverage, this branch's remote replay ordering proof, #7133's merged alt-only-clear tests, and #7173's merged ordered-seq interleaving and session-revival tests on main | Normal-buffer clear semantics and metadata-only replay still need deterministic lower-layer proof; the merged #7133/#7173 tests are unregistered until rebase | Add `terminal-output.scrollback-restore` as a dirty-state exactness contract, seeded with #7133's and #7173's tests |
+| Renderer/mirror parser parity | #7148 (merged 2026-07-02) fixed the first known width divergence; its red-green width test exists on main | Parser configs can still drift silently on any axis besides widths; no parity gate command is registered | Add `terminal-mirror.parser-parity` built on the shared `terminal-unicode-provider.ts`/`pane-terminal-options.ts` seams |
+| Post-restore convergence detection in production | Anomaly-breadcrumb machinery exists in this branch | Nothing measures renderer-vs-mirror divergence after real restores; #7133-class corruption stays silent until a user notices | Add `terminal-observability.restore-convergence-selfcheck` probe plus telemetry |
+| Stale pixels after reveal (render-level) | #7133's reveal hardening and the live repro harness's refresh-repair oracle | Buffer oracles are blind to buffer-clean, pixels-stale variants by definition | Add `terminal-render.pixel-refresh-repair` to the release-blocking terminal-rendering-golden suite |
+
+## Convergence Oracles And Escape Detection
+
+The #7133 saga is the design input for this section. One user-visible symptom — stale terminal frames on worktree return — decomposed into five distinct mechanisms: the skipped alt-screen clear (#7133 primary fix), the WebGL render-model fossil (#7133 hardening), the daemon mirror unicode width divergence (#7148), the stale hidden-chunk ordering hole with seq-restart data loss (#7173), and desktop resizes never reaching the runtime mirror (#7192). None of the three was predictable in advance, and seventeen faithful fresh-window repro attempts missed the class because the bugs only exist in dirty, long-lived pane states. Mechanism-enumeration testing cannot catch this family. Layer-convergence assertion can.
+
+Orca holds four copies of what a terminal shows: the PTY application's intended screen, the main-process headless mirror buffer, the renderer xterm buffer, and the WebGL-rendered pixels. Every bug in this family is two adjacent copies silently disagreeing. The strategy is two layers that map to the program's two goals — catching issues nobody predicted, and catching regressions to issues already fixed.
+
+Catch novel escapes:
+
+1. `terminal-observability.restore-convergence-selfcheck`: after every hidden-to-visible reveal settles — whether or not a restore was triggered — compare a bounded row-hash sample of the renderer buffer against the current main-mirror state; on mismatch, record a compact content-free `terminal_lifecycle_anomaly` breadcrumb (machinery already in this branch) and count it in telemetry. Per-reveal scope is load-bearing: #7173's frozen-output variant restores faithfully and then silently drops later hidden output, so a restore-scoped probe would never fire. This turns silent corruption into a measurable signal — #7133 and #7173 would have appeared in anomaly counts weeks before the user reports — and after each fix the anomaly rate is the in-production regression detector. Budget: one bounded comparison per reveal, no polling, no content capture beyond hashed row samples.
+2. `terminal-render.pixel-refresh-repair`: the render-level variant is invisible to every buffer oracle by definition (buffer clean, pixels stale). The content-independent oracle: screenshot the revealed pane, force a full model-invalidating repaint, screenshot again; a material diff means something rendered stale, whatever the mechanism. Productize the existing live repro harness with this oracle plus a JS-level render-model-vs-buffer comparison, and land it in the release-blocking `terminal-rendering-golden` suite that `release-cut.yml` already runs against every release tag. Use nightly runs only as the interim phase to stabilize the per-platform diff threshold; golden-set membership is what makes "the next release does not ship this class" directly checkable.
+3. Session-corpus replay: record real agent-session byte streams (a Claude run is exactly the #7133 shape — banner, prompt echo, final frame with blank regions) as fixtures and replay them through the hide/skip/restore machinery under the convergence oracles. Real corpora contain escape-sequence conjunctions nobody writes by hand. This is fixture strategy for the gates above, not a separate gate.
+
+Catch regressions to known fixes:
+
+4. `terminal-mirror.parser-parity`: feed identical byte corpora to the renderer xterm configuration and the main headless mirror and assert cell-identical buffers, with both parsers constructed from one shared configuration module. #7148's divergence (Unicode 11 + ZWJ in the renderer, default width tables in the mirror) existed since the mirror was born and was catchable by a millisecond-scale unit test; this gate also blocks the next one-sided config drift.
+5. The dirty-state restore exactness contract as `terminal-output.scrollback-restore`'s command: apply snapshots onto adversarially dirty panes and assert the resulting buffer exactly equals the snapshot frame. Exactness is load-bearing — presence-of-marker oracles pass on merged frames. Seeded with #7133's merged alt-only-clear tests and #7148's positioned-overwrite width oracle.
+6. Manifest registration of every escape: each fixed escape's red-green test becomes gate evidence, and per the Operating Rules the fix PR must also strengthen the relevant convergence oracle, so each escape permanently shrinks the space for the next one.
+
+Seam placement is part of the design: convergence must be asserted at every adjacent pair of copies, not at one seam. #7192 proved this the hard way — a renderer-vs-mirror check is blind to corruption inside the mirror itself, because the restore is perfectly faithful to a corrupted source. That is why the geometry gate owns the runtime mirror's dimensions and reflow ordering as a first-class size authority, upstream of the seam this section's self-check watches.
+
+The honest limit stands: this shrinks the escape space rather than zeroing it. The first four mechanisms in the saga are caught by items 1, 2, 4, and 5 without predicting any of them; the fifth (#7192) escaped every renderer-side oracle as originally specced and was closed by adding the mirror to the geometry authority list — evidence that the escape-driven loop (item 6) is not optional. The three new gates and the strengthened `terminal-output.scrollback-restore` entry are registered in the manifest as `protection: "none"` gaps, so the checker, factory reviews, and this plan cannot claim them as coverage before their commands exist.
+
+## Immediate Fix Candidates
+
+These are the concrete pieces to prioritize before declaring the plan implemented:
+
+Three mechanical prerequisites gate everything below:
+
+- Materialize the split: commit each working-tree slice to its owning branch per the file assignment in the PR Split Plan, reset the diverged child branches (#7004, #7005, #7006, #7008) from the worktree versions of their files, and push. Until then, every evidence run in this plan and in the manifest refers to unreproducible state.
+- During the rebase, run the testFiles-vs-fresh-main audit and fix the confirmed #6949 WebGL rename by adopting `terminal-webgl-atlas-recovery.test.ts` and evaluating `pane-webgl-context-recovery.test.ts` and `pane-webgl-renderer.test.ts` for the containment gate.
+- Open the aggregated non-blocking `reliability-gates` CI job so promotion evidence starts accruing with machine provenance.
+
+1. Rebase or retarget the existing reliability PR stack onto fresh `origin/main`, then verify the stack still contains queued automatic resume claims and exact `hasPty` missing-session reconciliation. This branch has restored both deterministic slices, but final stack verification is still required.
+2. Extend the provider-session ownership test from the current queued/pending, bounded-index, same-session hook, and wrong-session hook proof to a real Electron repeated-activation test, plus bounded-work assertions for any new hook/status scans.
+3. Add or restore exact liveness tests: `hasPty(id) === false` tears down only the matching local PTY, while `true`, `null`, rejection, SSH/remote unknown states, and stale pre-bind responses do not close panes.
+4. Add a small derived terminal ownership snapshot helper for tests and diagnostics. It should report whether each live PTY is represented once by the intended tab/leaf and absent from stale maps after attach, reattach, exit, clear, hibernate, and restore.
+5. Keep the replay FIFO/burst fix and tests in the stack, rebased over #7133's alt-clear restore preamble in the same file. This branch now asserts that multiple replay notifications arriving during an async drain cannot overwrite one another, cannot reorder accepted replay snapshots, and cannot grow an unbounded replay queue under bursty snapshots; remaining replay work is normal-buffer clear semantics (alt-screen clear landed in #7133), metadata-only replay, and hidden/live-output interleaving.
+6. Wire `DaemonPtyAdapter.reconcileOnStartup`, or explicitly create a provider-contract gate and accepted gap if wiring is not safe yet.
+7. Keep the degraded daemon fail-closed contract in place and extend it to startup/restart coverage. This branch now prevents unknown existing ids from silently routing operation calls or restored worktree-scoped and legacy/non-scoped spawn calls to fallback; the remaining work is production startup reconcile, prior-worktree aliases, and live daemon-restart validation.
+8. Keep the daemon-router targeted-discovery fix in place. This branch now caches ownership discovered by `hasPty()` before later write/resize-style operations and fails closed for operations on unknown existing session ids, preventing legacy-daemon sessions from silently routing to the current daemon.
+9. Extend provider listing failure contract tests. This branch now asserts disconnected SSH relay ids are retained as deferred reconnect metadata and sidebar wake hints rather than attached PTY proof, a rejected SSH listing does not clear previously learned ownership, thrown SSH `hasPty` is unknown, provider expired attach does not silently fresh-spawn, deferred SSH passphrase cancellation does not reconnect, deferred attach uses saved session ids, transient deferred reattach failure preserves retry metadata instead of fresh-spawning, and expired deferred relay fallback clears stale pane/tab bindings before one fresh spawn. Remaining work is empty-but-ambiguous provider snapshots, WSL, remote-runtime mirror polling, live SSH relay timing, false-live visual state, pending-vs-attached UI status, and pane-close behavior.
+10. Extend the new lifecycle anomaly breadcrumbs into a bounded structured lifecycle trace outside E2E-only mode for reattach, expired sessions, provider-list failure, unknown ownership, fallback routing, replay, and terminal input recovery.
+11. Extend the hot-interaction listing count gate. The targeted `hasPty`, resize-on-resume, light tab/active-state resume, and closed Resource Manager polling slices are implemented; next it should prove `pty:listSessions()` calls are zero during raw focus, workspace switch, render, high-session Electron scenarios, and separately budget open status/diagnostic polling.
+12. Add a cheaper Resource Manager session-status path or an explicit budget for open-popover `pty:listSessions()` polling. The closed-popover poll is removed, but the visible Resource Manager inventory still fans out to providers and mutates ownership, so it should have a measured ceiling.
+13. Budget the new startup hydration counters for `hydrateLocalPtyRegistryAtBoot` in CI/runtime evidence, including repo/worktree enumeration count, git subprocess count proxy, daemon session-list duration, and confirmation that no startup-critical path awaits the scan.
+14. Extend the output backpressure proof. The daemon stream batcher now pauses after `write(false)`, prioritizes flush-immediate output from another session ahead of unrelated background backlog on `drain`, clears pure-backpressure queues on global cleanup, bounds queued stream bytes, and resumes on `drain`; main pending renderer output is capped per PTY and in total. Next, hidden-output floods and active key latency need metric artifacts under the perf budget.
+15. Add Windows platform gates rather than relying on Linux/macOS live PTY evidence: ConPTY liveness, keyboard reset, CJK repaint, shell resolution parity, and IME/native-text forwarding.
+16. Decide whether the legacy JSON runtime terminal subscribe path is supported. If supported, gate its snapshot/live-output byte caps; if not, mark it as an accepted gap with a deprecation path.
+17. Implement the dirty-state restore exactness contract as `terminal-output.scrollback-restore`'s first command, seeded with #7133's merged alt-only-clear tests and #7173's ordered-seq interleaving and session-revival tests: apply snapshots onto already-alt-screen panes with stale content in cells the new frame leaves blank, include revived sessions with restarted PTY seq counters in the dirty-state matrix, and assert exact buffer equality.
+18. Implement the parser-parity seed during the rebase, now that #7148 has merged: register `headless-emulator-unicode-width.test.ts` plus a shared parser-construction assertion over `terminal-unicode-provider.ts` and `pane-terminal-options.ts` as `terminal-mirror.parser-parity`'s first command. The test file exists on fresh main but not at this branch's stale merge-base, so the command cannot be registered before the rebase.
+19. Implement the per-reveal convergence self-check behind the existing anomaly-breadcrumb machinery — comparing the renderer buffer against current mirror state on every reveal, not only after restores, with a strict per-reveal budget and no content capture — and productize the stale-render repro harness into `terminal-render.pixel-refresh-repair` targeting the release-blocking terminal-rendering-golden suite.
+20. Register #7192's merged mirror-geometry red tests as the mirror-authority slice of `terminal-geometry.visible-convergence` during the rebase, covering both accepted-resize fan-out to the runtime mirror and reflow-vs-queued-output ordering.
+
+Next gates to add before claiming broad terminal reliability coverage:
+
+1. `terminal-platform.live-pty-liveness`: first focused Electron/live-PTY slice is implemented for local Linux/macOS-style terminals. One local macOS run asserts bind, single active PTY listing, focus, keyboard input, PTY echo, repeated workspace hide/restore of the same PTY, post-restore input, actual size propagation, and exit cleanup. Follow up with scrollback-after-restore, tab switch, app restart, and non-local providers before broad live-terminal claims; keep it `experimental` until it has flake/runtime and red-green evidence.
+2. `terminal-platform.windows-conpty-liveness`: Windows-specific gate for ConPTY spawn/activation, PowerShell resolution, cursor/rewrite repaint, CJK/IME bytes, and no stuck `stale_bootstrap`. Use lower-level provider contracts where possible and one focused Windows Electron smoke for the real ConPTY path.
+3. `terminal-input.windows-conpty-keyboard-reset`: Windows local ConPTY gate proving Enter, Backspace, Arrow, paste, and ordinary shell submission behave normally after an agent or TUI exits. Oracle should include PTY input logs proving stale Kitty/CSI-u mode is not still applied to standard keys.
+4. `terminal-render.windows-cjk-repaint`: Windows local ConPTY wide-glyph redraw gate. Oracle should combine xterm buffer/text evidence with screenshot/canvas/pixel evidence and a bounded refresh count.
+5. `terminal-shell.windows-resolution-parity`: local and daemon providers must resolve equivalent shell path, args, cwd, env, startup command delivery, and fallback behavior for PowerShell 5/7, cmd, Git Bash, WSL, and missing `pwsh`.
+6. `terminal-performance.input-throughput`: non-blocking perf gate for typing latency, hidden-output pressure, renderer CPU, terminal output throughput, resize churn, and store subscriber work. This should be metric-based, not screenshot-based.
+7. `terminal-performance.no-hot-list-sessions`: deterministic count gate proving `pty:listSessions()` is not called from typing, focus, tab/workspace switch, visibility resume, resize, render, or per-pane liveness paths.
+8. `terminal-performance.daemon-stream-backpressure`: provider/daemon contract gate proving daemon socket writes respect backpressure under output floods and do not starve active input.
+9. `terminal-performance.output-backpressure-budget`: main-to-renderer ACK and renderer scheduler budget gate. Suggested ceilings: peak renderer in-flight bytes <=8MB total, <=512KB per PTY plus active reserve, renderer queued chars <=2MB, dropped renderer backlogs 0 in normal perf scenarios, and hidden restore <=1000ms.
+10. `terminal-runtime.mobile-stream-budget`: runtime/mobile stream gate proving binary terminal snapshots stay <=512KB mobile or <=2MB requested, live pending output stays <=256KB while snapshot loads, chunks stay <=48KB, and batches stay <=64KB.
+11. `terminal-input.ime-and-synthetic-forwarding`: representative IME/input matrix with deterministic byte/cell oracles at the lowest reliable layer, plus platform-specific manual/soak coverage where real IMEs are required.
+12. `terminal-provider.remote-daemon-contract`: provider contract for daemon degraded spawn, SSH/remote unknown liveness, mirror polling, reconnect, and path/cwd authority.
+13. `terminal-provider.wsl-restore-contract`: Windows WSL provider contract proving host/guest cwd identity, shell launch args, startup command delivery, targeted liveness semantics, and restore ownership, followed by one focused live WSL smoke.
+14. `terminal-output.scrollback-restore`: deterministic scrollback/snapshot/replay gate for hidden output, restore, clearing, and no cross-tab or stale-output overlap. Seed it with #7133's merged assertions — the alt-only clear preamble written before alt-screen snapshot data (and no `3J`) in `pty-connection.test.ts`, plus the reveal-repaint ordering tests — then extend to normal-buffer clear semantics, metadata-only replay, and hidden/live-output interleaving. #7133's buffer-merge escape (pre-hide frame bleeding through blank cells) is the motivating regression for this gate.
+15. `terminal-capability.startup-color-query`: startup OSC 10/11 replies are answered out of band at startup only, do not leak into shell/renderer streams, and ordinary OSC color queries remain renderer-handled.
+16. `terminal-mirror.parser-parity`: renderer xterm and the main headless mirror parse identical byte corpora into cell-identical buffers, with both parsers constructed from one shared configuration module. Seed with #7148's merged width test (`headless-emulator-unicode-width.test.ts`) plus a shared-construction assertion over `terminal-unicode-provider.ts` and `pane-terminal-options.ts`; extend with recorded agent-session corpora.
+17. `terminal-observability.restore-convergence-selfcheck`: production per-reveal probe comparing a bounded row-hash sample of the renderer buffer against current main-mirror state — on every reveal, not only after restores, because #7173's frozen-output variant never triggers another restore — recording deduped content-free anomaly breadcrumbs and telemetry counts. This is the in-production escape detector for the whole restore family; corruption upstream of the mirror itself is owned by the geometry gate.
+18. `terminal-render.pixel-refresh-repair`: content-independent refresh-repair oracle (screenshot, force a model-invalidating repaint, screenshot, bounded diff) plus a render-model-vs-buffer comparison, productized from the existing stale-render repro harness into the release-blocking `terminal-rendering-golden` suite; nightly runs only until the diff threshold is stable.
+
+Specific near-term hardening before promotion:
+
+- #7008 now includes a bounded queued-claim assertion plus same-session and wrong-session hook cases for provider-session ownership checks. It should still be followed by a real Electron repeated-activation proof and any additional bounded-work assertions if hook/status ownership scans grow.
+- Each gate PR should keep its residual gaps in the PR description and manifest until a real follow-up gate removes them. Do not let a green lower-layer gate imply live Electron, SSH, WSL, Windows ConPTY, or IME coverage.
+
+Promotion rule for these additions is unchanged: do not make any of them blocking until they have red/green proof, stable runtime history, zero unexplained flakes in the promotion window, a clear owner, and a demotion rule. A non-blocking gate is useful evidence and documentation; it becomes protection only after promotion criteria are met.
+
+## Goal
+
+Prevent the recent terminal, tab, session, startup, provider, and performance regressions from escaping again by turning each high-risk reliability class into a small executable gate with explicit maturity, owner, runtime budget, flake history, and promotion evidence.
+
+This plan is broad in scope but staged in rollout. The first milestone exercises the operating model end to end; later milestones add live Electron, Windows ConPTY, SSH/remote, daemon, IME, scrollback, and performance gates without turning the suite into noisy sprawl.
+
+## Operating Rules
+
+- Add gates only when they protect a named invariant tied to a real issue, PR, or accepted gap.
+- Prefer deterministic unit or provider-contract tests over Electron UI tests when the lower layer proves the user-visible invariant.
+- Never promote a gate to blocking until it has red/green evidence, stable runtime evidence, and a named demotion rule.
+- Keep stress/torture runs non-blocking unless they have deterministic oracles and flake history.
+- Include a performance or crash-safety budget whenever the protected change touches terminal throughput, hidden panes, persistence, startup, polling, subprocesses, git/worktree scans, SSH, WSL, or Windows paths.
+- Restore, replay, reattach, and snapshot gates must start from adversarially dirty initial states — already-alt-screen panes, stale content occupying cells the new frame leaves blank, scrollback present, wide/emoji glyphs — never only fresh terminals. Fresh-state-only testing is why the #7133 class survived seventeen faithful repro attempts.
+- Restore and replay oracles must assert frame exactness or stale-content absence. Marker-presence oracles pass on merged frames and are not acceptable correctness evidence for restore paths.
+- Every escaped terminal bug's fix PR must ship both the narrow mechanism regression test and a strengthening of the relevant convergence oracle (parser parity, restore exactness, convergence self-check, or pixel refresh-repair), so each escape permanently shrinks the space for the next one.
+
+## Milestone 1
+
+Milestone 1 creates the reliability-gate loop with the two highest-leverage escaped classes:
+
+1. `terminal-session.snapshot-freshness`
+   A stale local/daemon liveness snapshot cannot close a newer PTY binding.
+
+2. `agent-session.provider-ownership`
+   Workspace activation, restore, sleep, hibernate, dedupe, clearing, or reconnect code cannot replay or resume a provider session already owned, queued, pending, or live in the workspace.
+
+Both gates already have targeted tests in the repo. Milestone 1 makes them reviewable factory artifacts by registering their invariant, command, owner, maturity, known gaps, and promotion criteria in `config/reliability-gates.jsonc`.
+
+## First Commands
+
+Run the structural manifest check:
+
+```sh
+pnpm run check:reliability-gates
+```
+
+Run the first two gate tests:
+
+```sh
+pnpm exec vitest run --config config/vitest.config.ts \
+  src/renderer/src/components/terminal-pane/terminal-dead-session-reconcile.test.ts \
+  src/renderer/src/lib/resume-sleeping-agent-session.test.ts
+```
+
+Run the existing terminal perf report gate before promoting terminal lifecycle gates to blocking:
+
+```sh
+pnpm run test:e2e:terminal-perf:scale:report
+```
+
+## Promotion Criteria
+
+A gate can move from `experimental` to `soak` when:
+
+- it is registered in the manifest;
+- it has a deterministic oracle with no blind sleeps;
+- it has a cheap command that can run repeatedly;
+- it names the failure artifact reviewers should inspect;
+- it has a clear owner and demotion rule.
+
+A gate can move from `soak` to `blocking` when:
+
+- it has red/green evidence against the old behavior, a regression fixture, or an intentionally broken variant;
+- it has at least 100 consecutive passing soak runs or 14 days of required-platform CI history, whichever is more appropriate for the gate;
+- that history comes from the aggregated `reliability-gates` CI job with machine-recorded results (see Evidence Provenance And CI Wiring), not from hand-entered local runs;
+- p95 runtime is within the manifest budget;
+- there are zero unexplained flakes in the promotion window;
+- any required perf/git-crash budget has measured evidence.
+
+For visual and live release-assurance gates, `blocking` means membership in the release-blocking `terminal-rendering-golden` suite that `release-cut.yml` runs against every release tag — the existing golden rail, not a new checkpoint mechanism, and never PR-blocking. Deterministic unit and contract gates stay on the PR path, where catching a regression is strictly cheaper than catching it at release time.
+
+The manifest also allows `accepted-gap` for explicitly deferred reliability work with a named owner and `deprecated` for gates superseded by a stronger invariant or removed product surface. Those levels should stay visible in the manifest so factory review can distinguish intentional gaps from missing coverage.
+
+## Factory Contract
+
+`brennan-yolo-lite`, `review-code`, `perf`, and `git-crash-perf` should treat the manifest as the source of truth.
+
+For any PR touching a P0-capable surface, the agent or reviewer should:
+
+- identify the touched reliability class;
+- name the matching manifest gate or accepted gap;
+- run the gate command or explain why it does not apply;
+- require a new manifest entry when the PR introduces a new reliability class;
+- invoke `perf` or `git-crash-perf` when the touched surface matches their risk scope.
+
+## Evidence Provenance And CI Wiring
+
+Hand-entered evidence is the weakest link in this plan: the checker validates metadata shape, but nothing yet proves a registered command actually ran against the code it claims to cover. Reliability gates become protection only when they run automatically, against pushed commits, with recorded history. Requirements:
+
+1. Every `evidenceRuns` entry must record the pushed commit SHA it ran against. A run against uncommitted working-tree state is a development note, not promotion evidence; it must be re-run and re-recorded once the owning PR's branch is pushed. All evidence runs currently in the manifest predate the split and must be refreshed this way.
+2. Add one aggregated non-blocking `reliability-gates` CI job that runs `check:reliability-gates` plus every command-backed gate command, and uploads a machine-readable per-gate result artifact (gate id, command, result, duration, commit SHA, runner platform). Run it nightly and on PRs that touch terminal/session/provider/startup paths. Because every gate command except the live Playwright slice is a focused vitest run measured in seconds (a combined run of all 28 unit-layer gate test files completed in 10.2s with 928 tests on 2026-07-02), the job's cost is dominated by setup, not by gate count.
+3. Promotion to `active`/blocking requires history from that CI job — the 100-consecutive-run or 14-day criteria in Promotion Criteria — not local runs. Flakes recorded by the job must be triaged before the affected gate can be promoted, and an unexplained flake on an `active` gate is the demotion signal.
+4. Checker follow-up: once the CI artifact exists, extend the checker so `active` gates require at least one CI-provenance evidence run, and so evidence-run commit SHAs are verified to be ancestors of the branch under check.
+5. Rebase audit: before any merge claim, mechanically verify every declared `testFiles` entry exists on fresh `origin/main` (or on the rebased tree) rather than trusting memory. The #6949 WebGL rename is the standing example of why.
+
+## Milestone 2
+
+After Milestone 1 works, add the next three gates as experimental entries with deterministic harnesses:
+
+- `terminal-geometry.visible-convergence`
+- `xterm-addon.boundary-containment`
+- `startup-upgrade.persisted-session-corpus`
+
+These should not become blocking until the team has red/green, runtime, and flake evidence. The goal is fewer useful gates, not a larger noisy suite.
+
+## Milestone 3
+
+Milestone 3 is where the plan stops being only a lower-layer regression net and starts exercising the real terminal runtime:
+
+- Register existing terminal perf and E2E tests as experimental manifest gates only after reviewing their oracles, waits, artifacts, and flake history.
+- Add the live local PTY gate with a tight oracle: spawn a shell, print a unique token, switch hidden-visible, assert the same PTY/session id, verify visible buffer content, type again, and prove no duplicate tab or blank pane appeared.
+- Add Windows ConPTY coverage on Windows CI for echo, resize/readback, cursor visibility, CJK/emoji output, keyboard reset, shell resolution parity, and activation after restore.
+- Add daemon degraded/startup reconcile contracts before relying on daemon fallback behavior.
+- Add live SSH, WSL, remote-runtime mirror polling, and provider-listing edge contracts before treating remote/SSH restore as broadly covered; the current mocked SSH deferred-reattach and expired-relay tests are useful lower-layer proof, not live remote coverage.
+- Add perf gates with explicit budgets for input latency, event-loop delay, hidden-output pressure, daemon stream backpressure, no hot `listSessions`, pending bytes, renderer ACK behavior, store hot paths, and runtime/mobile stream byte caps.
+- Add IME/synthetic-input coverage at the lowest deterministic layer, with platform-specific manual or soak coverage only where real IMEs are required.
+
+Milestone 3 gates should begin as `experimental`. They become `soak` only when the command is stable enough to run repeatedly and the failure artifact tells an engineer exactly what broke. They become `blocking` only after the promotion criteria above are satisfied.

docs/reference/reliability-pain-points-2026-06-30.md

@@ -0,0 +1,712 @@
+# Reliability Pain Points and Improvement Plan
+
+Date: 2026-06-30
+
+## Scope
+
+This note summarizes the recent reliability review across:
+
+- Recently merged reliability PRs, especially terminal/tab/startup/session work from 2026-06-26 through 2026-06-30.
+- GitHub issues labeled `P0` and `P1`.
+- Local git history for authorship when issue text or PR bodies named a regressing change.
+
+The goal is to identify where Orca is hurting, what engineering improvements would reduce repeat incidents, and what can fairly be said about attribution.
+
+## Issue Snapshot
+
+At the time of review:
+
+- Open `P0`: 9
+- Total `P0`: 14
+- Open `P1`: 191
+- Open `P1` split: 90 bugs, 86 enhancements, 15 other
+
+Last seven days reviewed: 2026-06-23 through 2026-06-30.
+
+- Merged PRs in the window: 386
+- `P0` issues created in the window: 8
+- `P1` issues created in the window: 83
+
+The last-seven-day issue set is the strongest signal for the current pain. It includes the new-tab autoclose `P0`, Windows terminal activation hangs, renderer CPU/input lag, Windows startup visibility failure, Codex config corruption on Windows, a large terminal/IME/rendering cluster, WSL/remote/provider drift, and agent/sidebar/session identity drift.
+
+Open `P0` issues:
+
+- [#6877](https://github.com/stablyai/orca/issues/6877): macOS 1.4.106 nested `simcam` dylib is unnotarized and rejected by Gatekeeper.
+- [#6874](https://github.com/stablyai/orca/issues/6874): Windows `AppHangB1` when opening or activating a terminal session; runtime stuck in `stale_bootstrap`.
+- [#6873](https://github.com/stablyai/orca/issues/6873): likely duplicate or near-duplicate of #6874, though the issue metadata says macOS while the title/body describe Windows.
+- [#6795](https://github.com/stablyai/orca/issues/6795): app became very slow after an update; typing can lag by 2-5 seconds.
+- [#6773](https://github.com/stablyai/orca/issues/6773): cannot create a new terminal tab or agent; the tab opens and closes immediately.
+- [#6655](https://github.com/stablyai/orca/issues/6655): high `Orca Helper (Renderer)` CPU on Mac Mini M4 causing severe UI/input lag.
+- [#5787](https://github.com/stablyai/orca/issues/5787): whole Windows window freezes; input dies in every pane; sessions close on click.
+- [#5356](https://github.com/stablyai/orca/issues/5356): session loss on first launch after upgrading from 1.4.65 to 1.4.68.
+- [#5314](https://github.com/stablyai/orca/issues/5314): Windows close button and Alt+F4 do nothing.
+
+Closed `P0` issues worth keeping in the reliability history:
+
+- [#6233](https://github.com/stablyai/orca/issues/6233): Windows window fails to appear while the process runs in the background.
+- [#6163](https://github.com/stablyai/orca/issues/6163): Orca-managed Codex `config.toml` duplicate key.
+- [#5377](https://github.com/stablyai/orca/issues/5377): native watcher main-process crash on macOS.
+- [#5144](https://github.com/stablyai/orca/issues/5144): Windows close-confirmation deadlock.
+- [#5109](https://github.com/stablyai/orca/issues/5109): sidebar agent status mismatch.
+
+## P1 Reliability Clusters
+
+The `P1` set is too large to treat as a flat list. The high-signal reliability clusters are:
+
+### Terminal Rendering and Input
+
+- [#6901](https://github.com/stablyai/orca/issues/6901): garbled text.
+- [#6764](https://github.com/stablyai/orca/issues/6764): Windows terminal issues.
+- [#6632](https://github.com/stablyai/orca/issues/6632): display artifacts in Claude Code full-screen TUI.
+- [#6144](https://github.com/stablyai/orca/issues/6144): laggy Claude Code terminal scroll.
+- [#5969](https://github.com/stablyai/orca/issues/5969): remote-over-SSH Codex display artifacts.
+- [#5345](https://github.com/stablyai/orca/issues/5345): Windows/WSL scrolling causes duplicated or garbled text and content overlap.
+
+### IME and International Text
+
+- [#6905](https://github.com/stablyai/orca/issues/6905): Vietnamese IME input broken in terminal.
+- [#6765](https://github.com/stablyai/orca/issues/6765): Linux Sogou Pinyin/fcitx only commits the first CJK character.
+- [#6698](https://github.com/stablyai/orca/issues/6698): Vietnamese Telex input loses characters.
+- [#5921](https://github.com/stablyai/orca/issues/5921): CJK characters duplicated in Orca terminal on Windows.
+- [#5262](https://github.com/stablyai/orca/issues/5262): Arabic text rendering appears reversed or disconnected in terminal input.
+
+### Paste and Clipboard
+
+- [#5365](https://github.com/stablyai/orca/issues/5365): terminal freezes for seconds when pasting long text on Windows.
+- [#5358](https://github.com/stablyai/orca/issues/5358): Windows paste preview collapses long text and cannot expand it.
+- [#5919](https://github.com/stablyai/orca/issues/5919): generic paste reliability issue.
+- [#5960](https://github.com/stablyai/orca/issues/5960): bad screen result when pasting links.
+- [#6364](https://github.com/stablyai/orca/issues/6364): pasting images into an agent CLI does not work using Remote Host.
+
+### SSH, Remote Runtime, WSL, and Provider Boundaries
+
+- [#6106](https://github.com/stablyai/orca/issues/6106): SSH terminal loses pre-TUI shell output after Codex tab restore.
+- [#6846](https://github.com/stablyai/orca/issues/6846): remote PTY sessions do not stay alive on disconnect.
+- [#6908](https://github.com/stablyai/orca/issues/6908): WSL project added through UI does not discover branches/base refs, while CLI repo add works.
+- [#6907](https://github.com/stablyai/orca/issues/6907): Codex launched in a WSL worktree does not appear in agent status/sidebar.
+- [#6916](https://github.com/stablyai/orca/issues/6916): OMP launched in a WSL worktree has no live agent status.
+- [#6032](https://github.com/stablyai/orca/issues/6032): Remote Orca Servers can lose host-scoped repo and project state.
+- [#6688](https://github.com/stablyai/orca/issues/6688): viewing a diff on remote host errors.
+- [#6753](https://github.com/stablyai/orca/issues/6753): remote-host GitHub merge errors.
+
+### Agent, Tab, Session, and Sidebar State Drift
+
+- [#6910](https://github.com/stablyai/orca/issues/6910): active agent tabs duplicate UI sessions when switching repositories.
+- [#6803](https://github.com/stablyai/orca/issues/6803): multiple working agents thrash their order in the sidebar.
+- [#6072](https://github.com/stablyai/orca/issues/6072): mobile keeps showing old agent rows after terminals are closed.
+- [#5913](https://github.com/stablyai/orca/issues/5913): Pi compact agent row remains after closing its terminal tab.
+- [#5718](https://github.com/stablyai/orca/issues/5718): OMP sessions not showing up in Agent Session History or filter options.
+- [#5404](https://github.com/stablyai/orca/issues/5404): false "agent running" notifications persist after starting Claude Agents on VSSH/WSL.
+
+### Mobile Terminal and Tabs
+
+- [#5421](https://github.com/stablyai/orca/issues/5421): iOS workspace tabs do not open.
+- [#6756](https://github.com/stablyai/orca/issues/6756): iOS terminal renders in cursive/italic font and continuously flickers.
+- [#5628](https://github.com/stablyai/orca/issues/5628): mobile resizing feature request, closely related to terminal fit/reflow reliability.
+
+### Browser, Profile, and Embedded Runtime
+
+- [#6923](https://github.com/stablyai/orca/issues/6923): browser profiles do not isolate storage.
+- [#6760](https://github.com/stablyai/orca/issues/6760): floating browser cannot load any page.
+- [#6268](https://github.com/stablyai/orca/issues/6268): embedded-browser cookie/session limitation prevents Auth0 login.
+- [#6875](https://github.com/stablyai/orca/issues/6875): cannot copy Google Chrome cookies database.
+- [#6652](https://github.com/stablyai/orca/issues/6652): browser zoom broken.
+
+### Startup, Install, and Platform Hangs
+
+- [#5657](https://github.com/stablyai/orca/issues/5657): macOS startup PATH probe can hang under Endpoint Security agents.
+- [#5107](https://github.com/stablyai/orca/issues/5107): Orca opens very slowly on Windows 11 and leaves a residual process.
+- [#5989](https://github.com/stablyai/orca/issues/5989): Windows Defender blocks application during installation.
+
+## Recent Reliability PR Context
+
+The late-June reliability work clustered around the same few subsystems:
+
+- New terminal tabs opening and immediately closing: [#6796](https://github.com/stablyai/orca/pull/6796), [#6801](https://github.com/stablyai/orca/pull/6801).
+- Stale/frozen/blank terminals: [#6514](https://github.com/stablyai/orca/pull/6514), [#6800](https://github.com/stablyai/orca/pull/6800), [#6833](https://github.com/stablyai/orca/pull/6833), [#6830](https://github.com/stablyai/orca/pull/6830), [#6866](https://github.com/stablyai/orca/pull/6866).
+- PTY size and layout desync: [#6644](https://github.com/stablyai/orca/pull/6644), [#6649](https://github.com/stablyai/orca/pull/6649), [#6684](https://github.com/stablyai/orca/pull/6684), [#6725](https://github.com/stablyai/orca/pull/6725), [#6785](https://github.com/stablyai/orca/pull/6785), [#6853](https://github.com/stablyai/orca/pull/6853).
+- Renderer/xterm crash hardening: [#6852](https://github.com/stablyai/orca/pull/6852), [#6872](https://github.com/stablyai/orca/pull/6872), [#6855](https://github.com/stablyai/orca/pull/6855), [#6868](https://github.com/stablyai/orca/pull/6868), [#6856](https://github.com/stablyai/orca/pull/6856), [#6857](https://github.com/stablyai/orca/pull/6857).
+- Windows shell/spawn/input: [#6537](https://github.com/stablyai/orca/pull/6537), [#6876](https://github.com/stablyai/orca/pull/6876), [#6858](https://github.com/stablyai/orca/pull/6858), [#6890](https://github.com/stablyai/orca/pull/6890).
+- Hidden/agent startup behavior: [#6824](https://github.com/stablyai/orca/pull/6824), [#6798](https://github.com/stablyai/orca/pull/6798), [#6836](https://github.com/stablyai/orca/pull/6836).
+
+## Last-Seven-Day Regression Origin Map
+
+This section maps recent regressions to the place where the system failed, and to the gate that should catch the same class next time.
+
+| Issue | Symptom | Regression point | Fix / related PR | Missing gate |
+| --- | --- | --- | --- | --- |
+| [#6773](https://github.com/stablyai/orca/issues/6773) | New terminal/agent tab opens and closes immediately | Dead-session reconciliation could treat a fresh PTY binding as absent from a stale `listSessions()` snapshot. This likely traces to the local dead-session reconcile path introduced for hidden exited panes in [#6514](https://github.com/stablyai/orca/pull/6514). | [#6796](https://github.com/stablyai/orca/pull/6796), [#6801](https://github.com/stablyai/orca/pull/6801) | Newborn terminal lifecycle test: create shell and agent tabs while liveness reconciliation is in flight; assert the tab cannot be closed by an older snapshot. |
+| [#5356](https://github.com/stablyai/orca/issues/5356) | First launch after upgrade loses active floating/agent terminal sessions | Cold-restore relied on state that only versions after the fix could write, so users upgrading from pre-fix versions had no migration source. Issue text explicitly names [#5234](https://github.com/stablyai/orca/pull/5234) and [#5240](https://github.com/stablyai/orca/pull/5240). | Not fully closed in current open `P0` set | Upgrade fixture test: boot current app with persisted state from the last affected production version and assert active terminal/agent sessions are preserved or visibly recoverable. |
+| [#5319](https://github.com/stablyai/orca/issues/5319) | Linux/Wayland terminals render but stop accepting keyboard/scroll input | Wayland-aware GPU safeguard was removed in [#1344](https://github.com/stablyai/orca/pull/1344), leaving eager GPU channel setup able to wedge terminal input on affected systems. | [#6557](https://github.com/stablyai/orca/pull/6557) | Platform GPU matrix test: Linux Wayland launch with terminal WebGL/input smoke and assertion that GPU flags preserve input, not only rendering. |
+| [#6233](https://github.com/stablyai/orca/issues/6233) | Windows process runs but no window appears | Startup window reveal depended on Electron `ready-to-show`; if that event stalled, there was no fallback reveal or user-visible failure. | [#6462](https://github.com/stablyai/orca/pull/6462) | Startup watchdog test: simulate missing `ready-to-show` on Windows and assert the app either reveals or exits with a clear recovery path. |
+| [#6163](https://github.com/stablyai/orca/issues/6163) | Orca-managed Codex `config.toml` fails with duplicate `hooks.state` keys on Windows | Two TOML writers serialized the same Windows path key differently, and Orca deduped textually rather than by decoded key. | [#6318](https://github.com/stablyai/orca/pull/6318), follow-up [#6388](https://github.com/stablyai/orca/pull/6388) bug-scan fix | Cross-writer config round-trip test: seed config with Codex-written literal Windows paths, run Orca hook sync twice, then parse with a TOML parser and assert one decoded key. |
+| [#6244](https://github.com/stablyai/orca/issues/6244) | Headless-launched automation terminal showed only a bash prompt until unmount/remount | Hidden/background automation worktrees were not mounted early enough for output/replay to hydrate on first view. | [#6568](https://github.com/stablyai/orca/pull/6568) | Headless automation mount test: launch background automation, wait for output, first-focus the tab, and assert output is already visible. |
+| [#6331](https://github.com/stablyai/orca/issues/6331) | WSL workspaces with UNC cwd hit `DaemonProtocolError` | Windows/WSL path handling crossed provider boundaries and daemon spawn assumptions. | [#6536](https://github.com/stablyai/orca/pull/6536) | WSL UNC provider contract: add/open/delete workspace whose root is `\\wsl.localhost\...`; assert daemon and file operations use the right host/path semantics. |
+| [#6336](https://github.com/stablyai/orca/issues/6336) | Monorepo subfolder import opened the repo root instead of the selected subfolder | Git-root normalization leaked into the first-terminal cwd/user-facing workspace root behavior. | [#6574](https://github.com/stablyai/orca/pull/6574) | Add-project matrix: repo root, monorepo subfolder, folder workspace, remote runtime; assert file tree root and first terminal cwd separately. |
+| [#6147](https://github.com/stablyai/orca/issues/6147) | Chinese full-width punctuation became half-width ASCII in terminal | IME/composition forwarding lacked coverage for macOS full-width punctuation and composition-owned text. | [#6417](https://github.com/stablyai/orca/pull/6417) | IME matrix test: macOS CJK full-width punctuation, Vietnamese Telex, Linux fcitx/Sogou, Windows CJK; assert committed bytes and displayed cells. |
+| [#5656](https://github.com/stablyai/orca/issues/5656) / [#5653](https://github.com/stablyai/orca/issues/5653) | Windows Claude Code prompt showed phantom/overwritten characters until resize | Rewrite-style terminal output did not reliably force visible row repaint for CR/CHA/erase updates. | [#6544](https://github.com/stablyai/orca/pull/6544), [#6449](https://github.com/stablyai/orca/pull/6449) | Terminal redraw fixture: feed CR, CHA, backspace, erase-line/screen and assert visible cells after each chunk without resize. |
+| [#5161](https://github.com/stablyai/orca/issues/5161) | PowerShell terminal failed to spawn with Windows error code 5 | Windows shell resolver handed ConPTY a bare `pwsh.exe`, which could resolve to an App Execution Alias stub. | [#6537](https://github.com/stablyai/orca/pull/6537), [#6876](https://github.com/stablyai/orca/pull/6876) | Windows shell resolution contract: auto and explicit PowerShell selections must resolve to real executables and skip WindowsApps aliases. |
+| [#6270](https://github.com/stablyai/orca/issues/6270) | Restoring agent sessions was not working for SSH/runtime use cases | Resume logic and runtime host restrictions did not cover SSH workspaces. | [#6685](https://github.com/stablyai/orca/pull/6685) | Agent resume provider matrix: local, daemon, SSH, remote runtime, Windows shell; assert resume command is queued on the configured host/provider. |
+| [#6852](https://github.com/stablyai/orca/pull/6852) / [#6872](https://github.com/stablyai/orca/pull/6872) | Terminal search could crash terminal surface on narrow viewports | xterm search decorations received invalid dimensions and threw through the terminal surface. | [#6852](https://github.com/stablyai/orca/pull/6852), [#6872](https://github.com/stablyai/orca/pull/6872) | xterm addon safety test: narrow viewport, zero/near-zero columns, wrapped matches, keyboard next/previous; assert addon errors are caught and pane remains usable. |
+| [#6855](https://github.com/stablyai/orca/pull/6855) | xterm web-link provider `RangeError` could kill the Windows renderer | Link provider ran synchronously inside xterm with no guard around pathological wrapped lines. | [#6855](https://github.com/stablyai/orca/pull/6855) | Addon boundary policy: all xterm addon callbacks are guarded and failure disables only that addon for that pane. |
+| [#6853](https://github.com/stablyai/orca/pull/6853) | Split-right PTY stayed at `0x0` and rendered a white screen | Deferred spawn measured an unlaid-out pane; post-spawn reconcile could terminate before forwarding a usable size. | [#6853](https://github.com/stablyai/orca/pull/6853) | Split spawn invariant: visible panes must never keep PTY size `0x0`; fallback size must be forwarded if measurement is unavailable. |
+
+The repeated pattern is not one person repeatedly making the same mistake. It is that Orca does not yet have enough executable contracts around lifecycle, provider boundaries, terminal sizing, xterm safety, platform shell behavior, and upgrade state.
+
+## PR Process Evidence
+
+Reading the PR descriptions for the confirmed or likely regression chains shows that the failure was usually not "no process." The failure was that the process accepted plausible local validation without requiring the invariant that would have protected the escaped user path.
+
+### #6514: Dead-Session Reconcile Introduced a Newborn-Tab Hazard
+
+[PR #6514](https://github.com/stablyai/orca/pull/6514) was a Brennan-authored reliability fix for frozen terminal panes after a backgrounded agent exits. The PR body shows a substantial process: design review, completeness verification, headline behavior status, perf audit, code-review loop, Electron validation, and targeted unit tests. The stated non-goal explicitly excluded SSH/remote reconciliation, and the implementation revalidated pane/PTY identity at apply time.
+
+What still got through:
+
+- The PR protected "stale dead pane should close" but did not name the dual invariant "a fresh or newly bound pane cannot be closed by a liveness snapshot requested before that binding existed."
+- The tests covered live/dead/remote/idempotency paths, but not the TOCTOU case where `listSessions()` is requested, a new PTY binds before the response returns, and the stale response omits the newborn id.
+- Reviewers accepted "healthy panes are untouched" as a semantic claim, but the executable gate did not prove that claim under delayed session enumeration.
+
+Skill implication:
+
+- `brennan-yolo-lite`, `auto-design-review-fix`, `review-code`, and `brennan-test-changes` all need to force lifecycle changes to state the inverse safety property, not only the intended cleanup behavior.
+- For terminal lifecycle changes, the review checklist should ask "what newer state can this async observation accidentally destroy?"
+
+### #5234 and #5240: Session-Restore Fixes Missed Old-Version Upgrade Fixtures
+
+[PR #5234](https://github.com/stablyai/orca/pull/5234) and [PR #5240](https://github.com/stablyai/orca/pull/5240) were Jinwoo-authored fixes for slow-startup daemon PTY restoration and quit-time agent session persistence. Their PR descriptions show AI review, negative controls, targeted tests, and high-quality e2e coverage against the newly modeled behavior.
+
+What still got through:
+
+- The tests proved the new behavior once the new data and startup gates existed.
+- They did not appear to boot the current app against serialized persisted state from the affected production versions before the fix.
+- The escaped class is an upgrade/migration class: a fix can be correct for state written after the fix while still losing users who upgrade from state written before it.
+
+Skill implication:
+
+- Startup, restore, daemon, and persistence PRs need an "old state corpus" gate: seed data from the last affected production version, not only state produced by current code.
+- Negative controls should include "old production fixture fails before fix and is preserved after fix" when the bug is upgrade-related.
+
+### #1344: GPU Rendering Parity Missed Linux Wayland Input Failure
+
+[PR #1344](https://github.com/stablyai/orca/pull/1344) replaced global GPU opt-ins with VS Code-style GPU startup flags and terminal GPU acceleration settings. The PR description cites reference behavior and local tests, but it does not show the later, stricter Brennan PR process markers used in the June reliability fixes.
+
+What still got through:
+
+- The change removed or bypassed the Wayland-aware GPU safeguard later identified by [PR #6557](https://github.com/stablyai/orca/pull/6557).
+- Validation covered configuration and pane lifecycle tests, but not a Linux Wayland terminal input smoke where rendering appears live while keyboard/scroll input is wedged.
+- One TypeScript command was reported as failing on existing project-include issues, which made the PR less clean as a release-risk artifact.
+
+Skill implication:
+
+- Cross-platform rendering/startup changes need an explicit platform matrix tied to the failure mode: not just "renders," but "renders and accepts input."
+- Reference implementations are useful context, but factory review must still require Orca-specific platform hazards to be named and tested or accepted.
+
+### Regression Lessons From Fresh Review
+
+Fresh reviewer passes on the recent regression chains pointed to these reusable misses. These are class-level gates, not new blame findings.
+
+- [#6514](https://github.com/stablyai/orca/pull/6514) -> [#6773](https://github.com/stablyai/orca/issues/6773), fixed by [#6796](https://github.com/stablyai/orca/pull/6796) and [#6801](https://github.com/stablyai/orca/pull/6801): the original PR correctly proved the forward cleanup case, where a hidden local/daemon PTY exits and should reconcile on resume while live listed panes survive. The missed invariant was the inverse lifecycle property: a destructive liveness snapshot can only close state that existed before the snapshot was requested. The escaped path was a macOS local new shell or agent tab where `listSessions()` was requested, a new PTY bound before the response returned, the stale response omitted the newborn id, and the renderer routed the healthy binding through exit teardown. The next gate is a Terminal Snapshot Freshness and Ownership Gate: a fake daemon holds `listSessions()` at `t0`, binds a newborn PTY at `t1`, resolves the stale `t0` snapshot without that PTY, and asserts the tab/pane survives, still owns the same PTY, accepts input, and renders output. A later post-bind snapshot may close a genuinely dead PTY, but only when provider, pane, PTY id, and binding epoch still match.
+- [#6800](https://github.com/stablyai/orca/pull/6800), with [#5240](https://github.com/stablyai/orca/pull/5240), [#6411](https://github.com/stablyai/orca/pull/6411), [#6514](https://github.com/stablyai/orca/pull/6514), and [#6833](https://github.com/stablyai/orca/pull/6833) as related context: the original fix correctly proved that active worktree-sleep records should not be claimed merely because old tab/layout state exists; visible connecting panes with restorable identity can own wake, stale hidden panes fresh-resume, and completed hibernation evidence stays passive. The missed invariant was provider-session ownership, not "pane connects now." For each provider-session claim key, exactly one durable owner must exist: a live hook/status for that same provider session, a queued/sent resume command for that same provider session, or an explicit retained record because spawn/connect failed. Layout match, PTY wake hints, replayed terminal bytes, and cold-restored screen snapshots are display evidence, not ownership evidence. The next gate is an Agent Provider Session Ownership Matrix covering active focused tabs, inactive tabs, visible non-focused splits, live/quit/worktree-sleep origins, replay-only/cold-restore-only/wrong-session/no-hook/delayed-hook cases, and repeated workspace activation.
+- [#5234](https://github.com/stablyai/orca/pull/5234) / [#5240](https://github.com/stablyai/orca/pull/5240) -> [#5356](https://github.com/stablyai/orca/issues/5356): the original PRs correctly proved slow-daemon restoration and newly written quit-origin provider-session persistence. The missed invariant was first-open-after-upgrade compatibility: startup/restore fixes must preserve or explicitly recover user sessions from the last affected production persisted-state schema, not only from state written by the fixed build. The escaped path was first launch after a 1.4.65 -> 1.4.68 upgrade, including Windows 10 active floating shell and Claude panes, where the old version had no `sleepingAgentSessionsByPaneKey` payload to consume. The next gate is a Persisted Session Upgrade/Restore Contract: boot current Orca against immutable user-data fixtures from the last affected version, including active and inactive tabs, visible and hidden splits, old tab-level PTY wake hints, missing `sleepingAgentSessionsByPaneKey`, live/quit/worktree-sleep/legacy origins, repeat activation, and a second restart after current code writes upgraded state. The oracle rejects blank replacement panes, duplicate resume tabs, silent session loss, and "works only after current code writes new state."
+- [#1344](https://github.com/stablyai/orca/pull/1344) -> [#5319](https://github.com/stablyai/orca/issues/5319) / [#6557](https://github.com/stablyai/orca/pull/6557): the original PR correctly attempted to replace broad GPU opt-ins with VS Code-style GPU channel flags and prove terminal GPU/WebGL fallback at unit level. The missed invariant was terminal platform liveness, not visual readiness. A terminal is healthy only if it renders output, receives focus, delivers keyboard input to the correct PTY, echoes process output back to the xterm buffer, scrolls, and survives GPU/renderer startup on the target platform. The next gate is a Terminal Platform Liveness Gate across Linux Wayland GPU paths, headless Weston plus at least one scheduled/manual headful Wayland run, control Linux/macOS/Windows paths when renderer policy is shared, active and inactive tabs, visible and hidden panes, first mount, tab switch, repeated workspace activation, local PTY, and SSH/remote PTY. Screenshots alone never pass.
+- [#6644](https://github.com/stablyai/orca/pull/6644) / [#6649](https://github.com/stablyai/orca/pull/6649) / [#6725](https://github.com/stablyai/orca/pull/6725) / [#6785](https://github.com/stablyai/orca/pull/6785) / [#6794](https://github.com/stablyai/orca/pull/6794) / [#6853](https://github.com/stablyai/orca/pull/6853) / [#6939](https://github.com/stablyai/orca/pull/6939): the fixes correctly proved pieces of the PTY geometry class: first-mount column desync, single-frame insufficiency, split-mount convergence, requested-size versus applied-size drift, mobile-owned PTY exceptions, visible split `0x0` fallback, and visible geometry reassertion after stable fit. The missed invariant was cross-layer terminal geometry authority. A visible desktop-owned terminal is not valid until xterm `cols/rows`, fit/proposed dimensions, applied PTY winsize, shell-visible `stty`/`stdout.columns`, and echo-wrap behavior converge after layout settles. Hidden/inactive panes may defer resize, and mobile/remote providers may intentionally diverge, but every transition into visible desktop ownership must converge. The next gate is a terminal geometry matrix with fault injection for first `0x0` fit, delayed layout settle, dropped/delayed resize acknowledgement, requested-size-versus-applied-size drift, SSH-style unknown applied size, hidden-to-visible activation, split create/collapse, restore, repeated workspace activation, and no redundant hidden SIGWINCH.
+- [#6852](https://github.com/stablyai/orca/pull/6852) / [#6872](https://github.com/stablyai/orca/pull/6872) / [#6855](https://github.com/stablyai/orca/pull/6855), with [#6868](https://github.com/stablyai/orca/pull/6868), [#6856](https://github.com/stablyai/orca/pull/6856), and [#6857](https://github.com/stablyai/orca/pull/6857) as nearby WebGL/renderer-risk hardening: the fixes correctly moved toward safe wrappers for search dimensions, keyboard search navigation, link providers, WebGL addon patches, and renderer/GPU crash fallback. The missed invariant was xterm addon boundary containment. Optional addons are untrusted boundary code: search decorations, link providers, WebGL attach/reset/dispose, renderer fallback, hover callbacks, and stale async callbacks must not throw across the pane, React, or window boundary. The next gate is an xterm-addon-boundary harness that injects throwing addon behavior under active tabs, inactive tabs, visible and hidden splits, hidden-to-visible resume, floating tab switch, repeated workspace activation, and pane destroy/recreate; the oracle is pane-scoped addon disable/degrade with one breadcrumb, no React/window crash, and surviving focus, input, PTY echo, and rendering.
+
+Concrete gate shapes from the fresh passes:
+
+- **Snapshot freshness harness**: defer and reorder `listSessions()` / `listTerminals()` responses. Request a snapshot at `t0`, bind a new PTY at `t1`, resolve the `t0` response without the new PTY, and assert the pane survives with PTY id, tab ownership, and input/output proof. Then request a fresh `t2 > t1` snapshot for a genuinely killed PTY and assert teardown works. Run active/inactive tabs, visible/hidden splits, first-input probes, restore/reattach, repeat activation, and SSH/remote unknown-liveness variants.
+- **Provider-session ownership matrix**: model provider-session ownership separately from terminal render output. For live, quit, active worktree-sleep, and completed worktree-sleep origins, prove each provider session is exactly one of owned, resuming, or intentionally retained. Cover active focused tabs, inactive hidden tabs, visible non-focused splits, delayed/missing/wrong hooks, duplicate legacy records, fresh-spawn failure, and repeat activation. Displayed replay must never count as ownership.
+- **Old-version restore corpus**: keep immutable user-data fixtures from the last affected production versions, especially pre-#5232 session state. Boot current Orca against those fixtures and assert every persisted PTY/agent record is preserved, resumed once, warm-reattached once, or visibly marked unrecoverable. Include second restart after current Orca writes upgraded state.
+- **Wayland platform liveness**: extend the Linux Wayland verifier so it proves terminal render, focus, keyboard input, PTY echo, scrollback movement, and GPU crash/stall absence across active/inactive tabs, visible/hidden splits, first mount, tab switch, restore, repeated activation, local PTY, and SSH/remote runtime. Screenshot-only evidence is not enough.
+- **Terminal geometry convergence**: add an `assertTerminalGeometryConverged` oracle that compares xterm cols/rows, fit/proposed dimensions, applied PTY size when available, shell-visible `stty size` / `$COLUMNS`, and deterministic echo/wrap behavior. Inject dropped resize, delayed resize until visible, stale applied size, null SSH-style readback, initial `0x0`, and layout settling after more frames than old fixed budgets.
+- **xterm addon boundary harness**: inject throwing search decorations, link providers, WebGL attach/reset/dispose, and stale async callbacks under active tab, inactive tab, visible split, hidden split, hidden-to-visible resume, floating tab switch, repeat activation, and pane destroy/recreate. Assert no `window.onerror` or React boundary hit, one structured breadcrumb, addon-scoped degradation, focus/input survival, PTY/output rendering, and WebGL fallback to DOM when needed.
+
+### Corrective PRs Show the Desired Direction
+
+The later fixes, especially [#6801](https://github.com/stablyai/orca/pull/6801), show the right shape: root-cause description, happens-before proof, explicit `auto-design-review-fix`, completeness verification, code-review loop, `brennan-test-changes`, and targeted tests for the stale snapshot/newborn binding race. [#6852](https://github.com/stablyai/orca/pull/6852), [#6853](https://github.com/stablyai/orca/pull/6853), and [#6855](https://github.com/stablyai/orca/pull/6855) also show good local invariant tests around xterm addon throws and PTY `0x0` sizing.
+
+The process improvement is to make that shape mandatory before the regression escapes, and to promote only the deterministic, red/green parts into blocking CI.
+
+## Core Pain Points
+
+### 1. Terminal Lifecycle Is Inferred Instead of Owned
+
+Many paths infer terminal state from a combination of pane visibility, provider session membership, stored pane layout, active leaf ids, sleeping-agent records, and xterm mount state. That makes it easy for a stale observation to tear down a fresh pane, or for a dead pane to look live.
+
+Common symptoms:
+
+- New tab opens and immediately closes.
+- Hidden terminal exits but the visible pane later looks frozen and swallows input.
+- Sleeping or hibernated agents get stranded in stale panes.
+- Mobile and desktop disagree about which terminal/session still exists.
+
+### 2. PTY Size Authority Is Split Across Too Many Places
+
+`FitAddon`, `ResizeObserver`, deferred `requestAnimationFrame`, mobile-fit locks, hidden-pane guards, PTY locks, split layout equalization, and provider resize forwarding all compete to decide the active terminal size.
+
+Common symptoms:
+
+- PTY born at `0x0`.
+- White split pane.
+- First-mount column desync.
+- TUI output wraps at the wrong width until manual resize.
+- Mobile-size terminals fight desktop-size restoration.
+
+### 3. xterm Addons Are Unsafe Boundary Code
+
+The terminal surface trusts addon behavior too much. Recent failures came from search decorations, web-link providers, WebGL texture atlas resets, CJK/wide glyph rendering, and synchronized-output buffering.
+
+Any addon that can throw synchronously or produce invalid dimensions can take down the workbench unless Orca wraps it defensively.
+
+### 4. Provider Semantics Drift
+
+Local PTY, daemon PTY, SSH relay, remote runtime, WSL, Windows ConPTY, Linux Wayland, and mobile sessions do not share the same truth source for liveness, size, replay, cwd, shell, or reconnect behavior.
+
+Common symptoms:
+
+- Local-only liveness checks are accidentally applied to SSH/remote panes.
+- Remote paths miss fixes made only in the renderer.
+- WSL works through CLI add but not UI add.
+- Daemon can be protocol-healthy while PTY spawn is degraded.
+
+### 5. Agent Identity Is Coupled to Terminal UI Details
+
+Agent status, sidebar rows, compact rows, mobile session tabs, sleeping-agent records, terminal helper identity, and pane reuse all share responsibility for "what agent is this?"
+
+Common symptoms:
+
+- Duplicate agent tabs.
+- Stale sidebar rows after terminal close.
+- Wrong agent icon/title.
+- False running/completed notifications.
+- Worktree switching changes agent ordering or ownership.
+
+Class-level invariant:
+
+- Agent/provider session replay must be ownership-safe. When code touches agent launch, restore, sleep, hibernate, dedupe, clearing, or workspace activation, it must prove Orca does not replay or resume a provider session id that is already owned, queued, pending, or live in that workspace. This applies across active and inactive tabs, visible and hidden panes, repeat activation of the same workspace, and session records whose origin is live, quit, or worktree sleep.
+- Motivating example: the #6800 escape showed how code can conflate "this pane will connect right now" with "this pane/session owns this provider session." A valid inactive or hidden existing session can then look unowned, causing workspace activation to replay `agent resume <same-session-id>` instead of recognizing the session is already represented. The gate should protect the whole class, not only the #6800 path.
+- Display evidence is not ownership evidence. Replayed terminal bytes, cold-restored screen snapshots, tab layout matches, and "will connect soon" state must never count as proof that a provider session id is already claimed.
+
+### 6. Upgrade and Startup Are Under-Tested as First-Class Flows
+
+Several failures only appear after an app update, daemon survival across versions, old persisted state, stale helper paths, or a first launch after migration.
+
+Common symptoms:
+
+- First launch after upgrade loses sessions.
+- Daemon remains reachable but cannot spawn PTYs.
+- App process runs but no window appears.
+- Release packaging/notarization works at top level but fails for a nested binary.
+
+### 7. Reliability Fixes Can Regress Performance or Crash Safety
+
+Several reliability fixes add watchers, reconciliation loops, terminal replay, git scans, or richer diagnostics. Those are high-risk performance surfaces if they wake hidden panes, serialize large state, multiply subprocesses, or make cleanup scans wait for final results before showing progress.
+
+Common symptoms:
+
+- Terminal input becomes correct but slower under foreground TUI redraw or background PTY output.
+- Hidden panes keep waking the renderer for data/title/spinner frames.
+- Store subscribers or persistence writes serialize large terminal/worktree maps on common interactions.
+- Git/worktree cleanup opens a modal that spins, duplicates rows, hangs on a subprocess, or crashes on Windows/WSL/SSH paths.
+- Safety checks run unbounded git subprocesses or trust stale cleanup evidence before destructive removal.
+
+## Proposed Improvements
+
+### A. Build a Terminal Lifecycle State Machine
+
+Create a single source of truth for terminal lifecycle:
+
+- `newborn`
+- `connecting`
+- `live`
+- `hidden-live`
+- `exited-observed`
+- `exited-unobserved`
+- `hibernated`
+- `restoring`
+- `degraded`
+- `closed`
+
+All teardown, restore, reconnect, hibernation, and liveness decisions should transition through this machine. Avoid having individual React effects independently infer lifecycle from partial state.
+
+### B. Make Liveness Reconciliation Epoch-Based
+
+Every pane/PTY binding should carry a monotonic renderer-local epoch or generation. Any async liveness snapshot must include the epoch it was requested against.
+
+Rules:
+
+- A snapshot requested before a bind cannot close that bind.
+- A provider membership result can only prove death when the current binding still exactly matches the request context: pane id, provider id, connection id, PTY id, and epoch.
+- Remote/SSH/WSL providers must supply provider-scoped liveness, or the reconciler must treat them as unknown.
+
+This generalizes the timestamp guard added for newborn terminal tabs and makes it harder to reintroduce the class.
+
+### C. Define a PTY Size Ownership Protocol
+
+Document and enforce one protocol for size authority:
+
+- Initial spawn size source.
+- Hidden-pane spawn behavior.
+- Mobile-fit lock behavior.
+- Split layout settle behavior.
+- Fallback size behavior.
+- Provider resize acknowledgement.
+- Applied PTY size reporting.
+- Dropped, delayed, or skipped resize recovery.
+- Hand-off from reconcile loop to live `ResizeObserver`.
+
+Add invariant tests that a visible terminal never remains `0x0`, and that xterm `cols/rows` converge with PTY-reported `cols/rows` under split, restore, reload, and mobile-fit scenarios.
+
+Class-level invariant:
+
+- A visible terminal pane must have one authoritative size path from layout measurement to xterm fit to provider resize to applied PTY size. `0x0` and unknown layout measurements are non-authoritative for visible panes. If a resize is skipped because of a mobile lock, PTY lock, hidden state, split layout settle, first mount, provider delay, or dropped acknowledgement, Orca must keep reconciling until the applied PTY size and xterm size converge or the pane enters an explicit degraded state.
+
+Required deterministic hooks:
+
+- Force first measurement to `0x0` during split-right spawn.
+- Delay split layout settlement across frames.
+- Hold and release mobile-fit or PTY locks while reconcile is active.
+- Drop or delay one provider resize acknowledgement.
+- Report applied PTY size separately from requested size.
+- Switch hidden-to-visible or worktree-to-worktree while a resize is pending.
+
+Required oracle:
+
+- After the fault, assert visible xterm `cols/rows`, provider-reported applied PTY `cols/rows`, and process-observed `stdout.columns/rows` converge.
+- Assert a width-sensitive TUI or long-line fixture renders without one-column wrapping.
+- Assert typed input reaches the process and echoed output appears after convergence.
+- Assert no test is promoted to blocking if it relies on blind `waitForTimeout()` instead of size/lifecycle events or controlled fault release.
+
+### D. Add Provider Contract Tests
+
+Start with a deterministic fake/fault-injection provider, then run the same lifecycle contract against real providers.
+
+The fake provider should be able to:
+
+- Delay and reorder `listSessions()` responses.
+- Return `unknown` separately from empty/live/dead.
+- Delay attach/replay and resize acknowledgements.
+- Kill processes while panes are hidden.
+- Drop or duplicate exit/replay/resize events.
+- Simulate stale cwd/helper paths.
+
+After the fake provider is stable, run the contract against:
+
+- local PTY
+- daemon PTY
+- degraded daemon wrapper
+- SSH relay
+- remote runtime
+- WSL path/shell combinations
+- mobile session replay
+
+Required scenarios:
+
+- spawn
+- hidden spawn
+- visible restore
+- hidden exit
+- reconnect
+- resize before/after mount
+- cwd deleted
+- helper binary moved after upgrade
+- process dies while renderer is hidden
+
+### E. Create a Terminal Torture Harness
+
+Add a nightly and pre-release reliability suite that exercises:
+
+- rapid tab create/close
+- agent create/close
+- split right/down
+- reload with split layout
+- hidden-to-visible switching
+- huge streaming TUI output
+- CJK, Vietnamese, Arabic, and emoji input/output
+- long paste, link paste, image paste
+- browser/terminal focus handoffs
+- hibernation wake
+- daemon restart/degraded fallback
+- mobile connect/disconnect during active terminal traffic
+
+This should record screenshots, terminal snapshots, PTY sizes, lifecycle transitions, and renderer CPU.
+
+Each scenario must also declare its oracle before it can be promoted beyond `experimental`: the observable pass condition, the failure artifact, and whether it is a discovery/stress run or a merge-blocking gate. Stress scenarios without deterministic oracles should stay nightly/pre-release only.
+
+### F. Wrap xterm Addons Behind Safe Adapters
+
+Search, links, WebGL, decorations, and input-protocol helpers should be treated as untrusted code:
+
+- Catch synchronous addon throws.
+- Validate all rows/cols/decoration dimensions before calling xterm APIs.
+- Disable or degrade an addon for a pane after repeated failures.
+- Emit a structured breadcrumb with pane id, provider, terminal size, addon name, and sanitized error.
+
+Class-level invariant:
+
+- xterm addon failures must be contained to the addon and pane. A search, link, WebGL, decoration, input-protocol, or keyboard-navigation error must never unmount the terminal surface, crash the renderer, break focus, or stop typed input/output for the PTY. Invalid dimensions, wrapped-line ranges, zero columns, texture atlas resets, and synchronous callback throws should either be validated away before the addon call or caught and downgraded after the call.
+
+Required deterministic hooks:
+
+- Force zero or near-zero terminal columns.
+- Force wrapped search and link ranges across line boundaries.
+- Throw from search, link, decoration, and WebGL callback boundaries.
+- Simulate WebGL texture/atlas reset or addon initialization failure.
+- Invoke next/previous search navigation while the search addon reports an invalid match.
+
+Required oracle:
+
+- The terminal pane remains mounted and focused.
+- The failing addon is disabled or downgraded only for the affected pane.
+- A structured breadcrumb names the addon and sanitized failure.
+- Typed input reaches the process and echoed output appears after the addon failure.
+
+### G. Add Upgrade Fixtures
+
+Keep serialized workspaces and daemon/session state from known old versions:
+
+- pre-#5232 session restore data
+- stale daemon cwd
+- stale node-pty helper path
+- old hibernated agent records
+- old mobile session snapshots
+- old Windows shell settings
+
+CI should boot the current app against these fixtures and assert no session loss, no blank first pane, and no startup deadlock.
+
+Fixture rules:
+
+- Name the exact production version and build source.
+- Store the pre-migration state immutably and copy it to a temp profile before boot.
+- Document the expected pre-migration shape and post-migration behavior.
+- Include stale daemon/helper/path state when that is the real user upgrade path.
+- Treat silent blank replacement as a failure. If exact restoration is impossible, the accepted outcome must be an explicit recoverable/degraded state or user-facing migration notice, not silent session loss.
+
+### H. Add Platform Startup, Input, and Release Artifact Contracts
+
+Some failures look visually healthy while the product is broken. Platform contracts must prove the user action, not only the frame.
+
+Required contracts:
+
+- Terminal rendering/input: on Linux Wayland, Windows ConPTY, and macOS, assert the terminal renders, accepts typed input, echoes process output, scrolls, and preserves focus after GPU/renderer startup paths touched by the PR.
+- IME/input: when terminal input handling changes, assert representative CJK, Vietnamese, Arabic, paste, and modifier paths at the lowest deterministic layer available.
+- Startup/window lifecycle: packaged app reveals, close button and Alt+F4 work on Windows, startup watchdogs recover when `ready-to-show` stalls, and PATH/shell probes time out with a recoverable state.
+- Release artifact: packaged builds include required nested binaries and pass signing/notarization checks, not only top-level app validation.
+- Daemon degradation: daemon reachable-but-unable-to-spawn and stale helper/cwd states produce a visible degraded path instead of blank or frozen terminals.
+
+### I. Add an Agent Identity Contract
+
+Provider and terminal tests are not enough for agent/session identity drift. Add a contract for:
+
+- Pane reuse and tab close.
+- Worktree switch and sidebar ordering.
+- Sleep/wake and hibernation records.
+- Provider session id capture and consumption.
+- Mobile rows and desktop rows converging after terminal close.
+- False running/completed notifications.
+- Agent history/filter visibility for restored sessions.
+- Provider session replay ownership: workspace activation, launch, restore, sleep/hibernate, dedupe, clearing, and reconnect code must not issue a second resume/launch for a provider session id already owned by an active tab, inactive tab, visible pane, hidden pane, queued pane, pending connection, or live session in the same workspace.
+- Session record origins: the same invariant must be proved for live sessions, quit-restored sessions, and worktree-sleep records, including repeat activation of an already-active workspace.
+
+Required agent identity scenarios:
+
+- Activate a workspace with an existing visible live agent pane; assert no duplicate `resume <provider-session-id>` is queued.
+- Activate a workspace with an existing hidden/inactive live agent pane; assert the provider session is recognized as represented even before the pane visibly reconnects.
+- Re-activate the same workspace repeatedly while agent restore/session discovery is pending; assert the pending ownership claim dedupes all replay attempts.
+- Restore a quit-origin session and a worktree-sleep-origin session; assert each is claimed once, survives hidden-to-visible transitions, and does not produce duplicate sidebar/mobile rows.
+- Clear or close stale panes while a same-session replacement is queued or pending; assert cleanup cannot erase the newer ownership claim or trigger a second replay.
+
+### J. Introduce Reliability Gates for Terminal/Session PRs
+
+For PRs touching terminal, agent status, sessions, startup, provider routing, or hibernation, require:
+
+- Provider matrix explicitly listed.
+- Hidden-pane behavior listed.
+- SSH/remote behavior listed.
+- Windows/WSL behavior listed when shell, cwd, path, ConPTY, or process launch is touched.
+- A regression test that fails against the old behavior.
+- A manual or automated stress command if the change touches timing.
+- PTY size authority proof when the PR touches terminal spawn, split layout, hidden/visible transitions, mobile fit, resize forwarding, PTY locks, provider resize acknowledgement, or xterm fit. The proof must cover `0x0` quarantine, requested versus applied PTY size, xterm/process size convergence, and a user-visible input/output oracle.
+- Agent/provider replay ownership proof when the PR touches agent launch, restore, sleep/hibernate, dedupe, clearing, workspace activation, pane reuse, provider session id capture, sidebar/mobile agent rows, or session history. The proof must cover active versus inactive tabs, visible versus hidden panes, repeat workspace activation, and live, quit, and worktree-sleep session record origins.
+- xterm addon containment proof when the PR touches search, links, WebGL, decorations, keyboard terminal shortcuts, synchronized output, or xterm addon registration. The proof must show addon throws and invalid geometry cannot crash the terminal surface, unmount the pane, or break typed input/output.
+- The terminal/session checklist should explicitly ask: "Can this change make an already represented provider session look unowned, or replay/resume it a second time?"
+- The terminal/session checklist should explicitly ask: "Can this change make a visible pane trust `0x0`, stale requested size, or renderer-only size instead of the size the PTY actually applied?"
+- The terminal/session checklist should explicitly ask: "Can an xterm addon callback or invalid addon geometry throw through the terminal surface instead of being contained?"
+
+### K. Make Reliability Tests Predictable and Useful
+
+The goal is not "more tests." The goal is a small set of executable contracts that protect named invariants from real failures.
+
+Required shape for every new reliability gate:
+
+- Name the invariant it protects.
+- Name the real issue, PR, or failure mode that motivated it.
+- Prove red/green: the gate must fail against the old behavior, a regression fixture, or a small intentionally broken variant before it becomes a trusted blocker.
+- Attach a promotion record: invariant id, motivating issue/PR, old-fail evidence, new-pass evidence, exact command, failure output or CI link, deterministic wait condition, owner, runtime budget, and demotion rule.
+- Pick the cheapest deterministic layer: pure state-machine or provider-contract tests first, integration tests second, Electron/UI tests only for the small number of golden paths that require the real app shell.
+- Avoid sleeps as proof. Wait for observable state, protocol events, lifecycle transitions, PTY size convergence, replay completion, or process exit.
+- Record why the test is allowed to block merges: expected runtime, flake history, covered invariant, and owner.
+
+Test maturity levels:
+
+- `experimental`: local or nightly only; useful for shaking out the harness, not allowed to block merges.
+- `soak`: runs repeatedly in CI but non-blocking; collects flake rate, runtime, and whether failures are actionable.
+- `blocking`: promoted only after red/green proof, stable runtime, and low flake rate.
+
+Flake policy:
+
+- Blocking promotion should require numeric evidence, for example 100 consecutive soak runs or 14 days across required CI platforms, p95 runtime within the gate budget, and zero unexplained flakes. Tune the numbers later, but do not leave "stable" undefined.
+- A blocking reliability test that flakes above the agreed threshold should be demoted or fixed quickly; rerun-only culture hides weak gates.
+- CI should quarantine or demote a blocking test automatically when it exceeds the threshold over the configured window, open or update an owner issue, and require a fresh soak window before re-promotion.
+- A flaky test should be treated as a product or harness bug with an owner, not as background noise.
+- If a scenario cannot be made deterministic yet, keep it as soak/manual evidence and do not let it become required CI until the signal is trustworthy.
+
+Deletion policy:
+
+- Delete or demote tests that do not map to an invariant, duplicate a stronger gate, assert implementation details instead of behavior, only fail from timing/environment noise, or require blind sleeps.
+- Prefer one high-signal contract test over several UI tests that all exercise the same happy path.
+- Track regression coverage by invariant, not by raw test count.
+- Maintain a reliability-gates manifest in the repo. Each gate should list invariant id, motivating issue/PR, layer, owner, maturity, runtime budget, flake history link, promotion record, and replacement gate if deprecated. Tests without current metadata should fail review or be demoted.
+
+How to know the program is working:
+
+- Every recent P0/P1 reliability class maps to an invariant and an executable gate or explicitly accepted gap.
+- New gates carry red/green evidence.
+- Blocking gate flake rate and runtime stay within budget.
+- Future incidents identify either a missing invariant, a failed gate, or an explicit accepted gap instead of requiring archaeology.
+- Accepted gaps have an owner, reason, affected platforms/providers, issue link, and expiry date. Expired gaps should block future high-risk PRs until renewed or closed.
+- The dashboard tracks touched high-risk PRs, gates required, gates implemented, accepted gaps by age/owner, promotions, demotions, and escaped regressions by invariant.
+
+### L. Make Performance and Git Crash Safety First-Class Gates
+
+Correctness gates must include a performance budget when the fix touches terminal throughput, hidden panes, store subscribers, persistence, startup, polling, git/worktree scans, cleanup modals, subprocesses, SSH/WSL providers, or Windows git paths.
+
+Required performance contract:
+
+- Name the user-visible performance symptom that could regress: typing latency, scroll/resize jank, terminal throughput, startup delay, modal loading, CPU, memory, subprocess churn, or crash/reload.
+- Name the resource being protected: renderer frame time, event-loop delay, PTY IPC count, xterm writes, serialized bytes, subprocess count, file descriptors, WebGL contexts, webviews, or retained buffers.
+- Add a before/after measurement or deterministic count/leak test. Do not rely on "felt faster."
+- Keep foreground input echo and control responses latency-sensitive while batching background or high-volume redraw streams.
+- Prove hidden snapshot-capable panes do not wake the renderer per PTY chunk/title frame and restore from main-owned buffers when shown.
+- Preserve `orca terminal read` pagination, cursor metadata, bounded preview behavior, partial-line handling, truncation flags, and total counts when changing terminal buffering.
+- Prove subscriber, persistence, and publication paths avoid unchanged large slices and record serialized byte size when persisted payloads are touched.
+- Keep stress/torture runs non-blocking until they have numeric runtime and flake evidence.
+
+Required git crash/perf contract:
+
+- Bound every git subprocess in scan/cleanup paths with a timeout or abort signal, and kill the process tree where applicable.
+- Use bounded concurrency and in-flight dedupe; avoid one subprocess per repo x worktree x polling tick.
+- Stream progress incrementally to the renderer and remove IPC listeners in `finally`.
+- Verify close/reopen/refresh during active or just-settled scans clears loading state, preserves rows, and does not duplicate progress.
+- Recheck cleanup candidates immediately before destructive delete, including dirty files, unpushed commits, unknown upstream/base, folder repos, main worktrees, pinned workspaces, running terminals, disconnected SSH, and WSL paths.
+- Keep local, WSL, and SSH git providers at parity for timeouts, aborts, errors, and path handling.
+- On Windows-risky cleanup changes, validate a real Electron flow when feasible: progress appears before final completion, close/reopen recovers, a clean old worktree delete does not crash, and no orphan Electron/git process remains.
+
+Promotion rule:
+
+- A reliability fix cannot become a blocking gate if it makes the product slower or crashier. The promotion record must include the perf command or measurement, budget, result, platform/provider scope, and any accepted perf gap.
+
+### M. Make the Factory Enforce Those Gates
+
+The factory should turn the reliability gate into default behavior for planning, review, and merge-confidence testing. This should be treated as part of the prevention plan, not as a substitute for the actual test harnesses and CI gates above.
+
+Factory changes to carry forward:
+
+- `brennan-yolo-lite` should require a Reliability regression gate in the design doc whenever terminal, tab, session, provider, startup, upgrade, persistence, packaging, shell, path, or release behavior is touched.
+- The gate should force the agent to classify touched P0-capable surfaces: terminal lifecycle, tab creation/binding, dead-session reconciliation, hidden-to-visible resume, PTY identity/routing, PTY geometry including `0x0` quarantine, xterm addon boundaries, IME/paste/redraw behavior, local/daemon/SSH/remote/WSL/mobile provider contracts, attach/replay/resize/exit ordering, startup/upgrade persisted state, daemon survival, renderer/GPU crash containment, packaging/notarization, shell/PATH probing, terminal throughput, hidden-pane output, store subscriber hot paths, persistence payload size, git/worktree scan subprocess churn, cleanup modal progress, and Windows/WSL/SSH git crash risk.
+- Completeness verification should block when a touched P0-capable surface has no implemented or explicitly accepted reliability gate.
+- `review-code` should treat missing reliability tests or validation gates as findings for these surfaces, with the same review weight as lifecycle, provider, or cross-platform regressions.
+- `review-code` should invoke the `perf` skill for slow UI, terminal throughput, high CPU, memory/resource leaks, polling, subprocess churn, startup latency, render churn, or persistence writes.
+- `review-code` should invoke the `git-crash-perf` skill for worktree cleanup, git scans/status/listing, destructive removal, cleanup modal progress, SSH/WSL git providers, subprocess timeouts, or Windows git crash risk.
+- `brennan-test-changes` should run or explicitly justify the relevant terminal torture, provider contract, startup/upgrade, persisted-state, release, cross-platform, perf, or git crash/perf matrix subset.
+
+Current implementation status:
+
+- Internal factory skill update PR: https://github.com/stablyai/orca-internal/pull/491
+- Scope of that PR: add the reliability gate to `brennan-yolo-lite` and expand `review-code` to inspect the recent terminal/session/provider/startup/release failure classes.
+- Follow-up needed: once the concrete repo-side harnesses exist, update the skills again to name the exact commands and required CI jobs instead of only naming the expected reliability evidence.
+
+Additional factory follow-up:
+
+- `auto-design-review-fix`: treat missing reliability gates for P0-capable surfaces as design findings; require inverse invariants for cleanup/reconcile/init code.
+- `review-code-fix-loop`: require the final loop report to name the touched reliability classes, invariant, gate, and whether both reviewers addressed them. A clean review that never discusses the required reliability class should not count as clean.
+- `brennan-test-changes`: add merge-confidence sub-gates for stale snapshot/newborn binding races, hidden-to-visible resume, attach/replay/exit ordering, old-version persisted state, provider/platform matrices, and render-live/input-live checks.
+- `pr-review`: classify P0-capable surfaces before review/fix and require the same invariant/test/platform matrix before approval.
+- `yolo-lite` and `electron`: require terminal/runtime validation to prove input, scroll, focus, process output, and relevant platform behavior, not only screenshots or visible frames.
+- `playwright-reliability-tests`: add a focused skill for writing and reviewing Playwright tests that are evidence-based, low-flake, and tied to named invariants, so browser/Electron tests do not become more noisy checklist coverage.
+- `perf`: require measurement-backed review for terminal throughput, hidden pane output, store subscribers, persistence, startup, polling, subprocess churn, resource leaks, and memory growth.
+- `git-crash-perf`: require bounded subprocesses, progress streaming, close/reopen recovery, fresh delete preflight, provider parity, and Windows Electron validation for git cleanup/worktree changes.
+
+### N. Add Pane Lifecycle Telemetry and Diagnostics
+
+Each terminal pane should be able to emit a compact lifecycle trace:
+
+- pane id, tab id, worktree id
+- provider and connection id
+- pty id and binding epoch
+- visibility transitions
+- spawn/connect timestamps
+- resize measurements and forwarded sizes
+- liveness snapshot request/response times
+- teardown reason
+- hibernation/restore reason
+
+This will make future reports much easier to attribute without guessing. It should also be test-consumable: CI should be able to fail on forbidden transitions, stale close attempts, stuck `0x0`, missing input echo, and startup watchdog expiry.
+
+## Attribution Methodology
+
+"Who caused it?" is not reliably answerable from labels alone. This review separates:
+
+- **Confirmed**: the issue body or fix PR explicitly names the introducing PR/change.
+- **Likely**: the fix PR identifies the failing subsystem and local git history shows the regressing mechanism was introduced by a specific PR.
+- **Area ownership**: the same people are repeatedly fixing or modifying the subsystem, but there is not enough evidence to say they caused a given issue.
+- **Systemic**: no single introducer is evident; the problem is a missing invariant, provider contract, or test harness.
+
+## Attribution Findings
+
+### Confirmed Regression Chains
+
+- [#5356](https://github.com/stablyai/orca/issues/5356), session loss on first launch after upgrading from 1.4.65 to 1.4.68, explicitly names PRs [#5234](https://github.com/stablyai/orca/pull/5234) and [#5240](https://github.com/stablyai/orca/pull/5240). Local git history shows both PRs were authored by Jinwoo Hong. Count: Jinwoo 1 confirmed open `P0` regression chain.
+- [#5319](https://github.com/stablyai/orca/issues/5319), Linux/Wayland terminal input freeze, is closed, but the fixing PR [#6557](https://github.com/stablyai/orca/pull/6557) says the Wayland-aware GPU path had been removed in [#1344](https://github.com/stablyai/orca/pull/1344). Local git history shows #1344 was authored by Neil. Count: Neil 1 confirmed closed `P1` regression chain.
+
+### Likely Regression Chains
+
+- [#6773](https://github.com/stablyai/orca/issues/6773), new terminal/agent tabs opening and immediately closing, was fixed by [#6796](https://github.com/stablyai/orca/pull/6796) and [#6801](https://github.com/stablyai/orca/pull/6801). Those PRs identify the dead-session reconciler as the cause. Local git history shows [#6514](https://github.com/stablyai/orca/pull/6514), which introduced the focused dead-session reconcile module for hidden exited panes, was authored by Brennan Benson. Count: Brennan 1 likely open `P0` regression chain.
+
+### Area Ownership Without Clear Causation
+
+The recent terminal reliability churn is concentrated among Brennan, Neil, and Jinwoo because they authored many of the fixes and nearby changes. That does not mean each issue was caused by them. It does mean the riskiest ownership surfaces are:
+
+- Terminal lifecycle/reconcile/session persistence: Brennan, Neil, Jinwoo.
+- PTY size/layout/reflow: Neil, Jinjing, Jinwoo, Brennan.
+- Daemon/degraded spawn/focus/remote lifecycle: Neil, Jinwoo.
+- Windows terminal/shell/ConPTY/GPU recovery: Brennan, Neil, Jinwoo.
+- Agent status/sidebar/session identity: Brennan, Neil, Jinwoo, Jinjing.
+
+### Current Conservative Count
+
+For currently open `P0`/`P1` issues with a clear or likely PR-caused chain:
+
+- Jinwoo: 1 confirmed open `P0` regression chain.
+- Brennan: 1 likely open `P0` regression chain.
+- Neil: 0 confirmed open `P0`/`P1` chains from the evidence pulled, plus 1 confirmed closed `P1` chain.
+
+Most open `P0`/`P1` issues do not identify an introducing PR. Treating those as individual blame would be misleading. The stronger conclusion is that Orca has systemic reliability gaps around terminal/session lifecycle, provider contracts, PTY sizing, and upgrade/startup flows.
+
+## Recommended Next Steps
+
+Implementation plan: [`docs/reference/reliability-gates-implementation-plan.md`](./reliability-gates-implementation-plan.md).
+
+1. Deduplicate and update `P0` issues #6873/#6874 and #6773 with the suspected fix status.
+2. Add narrow red/green gates for known escapes first: newborn liveness race, PTY `0x0`, xterm addon throw containment, old-version upgrade fixture, and Linux Wayland render-live/input-live smoke.
+3. Define test maturity levels, promotion records, numeric flake/demotion policy, and a reliability-gates manifest before making new reliability gates blocking.
+4. Build the fake/fault-injection provider contract kit, then run the same contract against local and daemon PTY before expanding to SSH, WSL, remote runtime, and mobile.
+5. Add upgrade/startup/release fixtures for pre-#5232 session data, stale daemon/helper paths, packaged app reveal/close, nested notarization, PATH probe timeout, and Windows terminal activation.
+6. Add the terminal lifecycle state machine design doc and implement it incrementally behind the known escaped-regression gates.
+7. Formalize the PTY size protocol and make `0x0` a quarantined/non-authoritative state for visible panes.
+8. Add the agent identity contract for pane reuse, worktree switch, sidebar/mobile rows, provider session ids, stale notifications, and ownership-safe provider session replay across active/inactive tabs, visible/hidden panes, repeated activation, and live/quit/worktree-sleep records.
+9. Add performance budgets and measurements to reliability gates that touch terminal throughput, hidden panes, store subscribers, persistence, startup, polling, subprocess churn, git/worktree scans, cleanup modals, SSH/WSL git providers, or Windows git crash risk.
+10. Gate terminal/session PRs behind the reliability checklist above, including perf and git crash/perf checks when those surfaces are touched.
+11. Keep the factory skill update as the enforcement layer, add the Playwright reliability-testing skill, and make `perf` plus `git-crash-perf` mandatory companion skills for relevant PRs; then revise companion skills once exact repo commands and CI job names exist.
+12. Add pane lifecycle traces so future issues can be attributed from evidence instead of archaeology.

docs/reference/ssh-typing-latency-under-relay-load.md

@@ -0,0 +1,115 @@
+# SSH Typing Latency Under Relay Load
+
+Date: 2026-07-06
+
+## Symptom
+
+Typing in an SSH-host terminal can become extremely slow (hundreds of ms to
+seconds per echoed keystroke) while other SSH work is active — file previews,
+source-control refresh, search. A quiet SSH shell stays snappy, which made the
+reports look unreproducible.
+
+This is distinct from the local warm-switch lag documented in
+`terminal-switch-typing-lag-investigation.md` (daemon `listSessions()`
+snapshot cost — already fixed; `TerminalHost.listSessions()` is now
+metadata-only via `getAppliedSize()`).
+
+## Root Cause
+
+The relay and the Electron client share ONE ordered SSH channel for all
+JSON-RPC traffic: PTY input/output, file streams, git responses, search, port
+scans. Two mechanisms turned bulk traffic into typing latency:
+
+1. **Relay-side head-of-line blocking (primary).** The `fs.readFileStream`
+   pump (`src/relay/fs-handler-file-read.ts`) wrote every 256KB chunk (~340KB
+   framed) into the relay's stdout as fast as local disk reads completed,
+   ignoring the `write() === false` backpressure signal. A 10MB preview
+   enqueued ~13.6MB into the pipe at once; a `pty.data` echo emitted
+   mid-stream queued behind ALL of it. At 2MB/s WAN that is multiple seconds
+   of echo delay per open file. The reproduced measurement: 4,195,592 bytes
+   (the entire remaining 3MB test file, framed) queued ahead of one echo
+   frame.
+
+2. **Client-side O(n²) frame buffering (secondary).** `FrameDecoder.feed()`
+   in both `src/main/ssh/relay-protocol.ts` and `src/relay/protocol.ts` did
+   `Buffer.concat([buffered, chunk])` per data event, re-copying the whole
+   backlog for every ~32KB TCP chunk — ~2MB of memcpy per 340KB frame,
+   ~80MB per 10MB file — on the Electron main thread, delaying `pty:write`
+   IPC dispatch and echo delivery to the renderer.
+
+## Fix
+
+- **Bulk lane with sink backpressure** (`RelayDispatcher.notifyBulk`):
+  `fs.streamChunk` frames are serialized per client and each send waits for
+  the sink's `drain` when `write()` returns `false`. Interactive frames
+  (`pty.data`) still use plain `notify()` and are admitted immediately —
+  because the bulk lane keeps the outbound buffer at ~1 frame, an echo jumps
+  ahead of every not-yet-admitted chunk. Sinks (`relay.ts` stdout + Unix
+  socket clients) surface `write()`'s boolean and a one-shot
+  `waitWriteDrain`; every death path (EPIPE, stdin end, detach, dispose)
+  flushes parked waiters so a pump can never hang on a dead pipe.
+
+- **Credit-based flow control** (`fs.streamAck`): the client requests
+  streams with `flowControl: 'ack'` and acks each processed chunk; the relay
+  caps unacked chunks at `STREAM_ACK_WINDOW_CHUNKS = 4` (~1MB raw). This
+  bounds in-flight bulk bytes even past the relay's own pipe (sshd/TCP
+  buffers) and paces the relay's base64/JSON encode loop so incoming
+  keystroke frames get event-loop turns. A parked pump wakes on ack, cancel,
+  release, client detach, and a 1s staleness recheck.
+
+- **Cross-version compatibility**: old client + new relay → no
+  `flowControl` param → legacy unpaced pump (still drain-bounded, which is
+  transparent). New client + old relay → `fs.streamAck` is an unknown
+  notification and is silently ignored. `STREAM_CHUNK_SIZE` (256KB) is
+  intentionally unchanged — both sides bake it into chunk offset math, so
+  changing it would corrupt cross-version streams.
+
+- **FrameDecoder rewrite** (both mirrored protocol files): chunk-list
+  buffering; each frame is assembled exactly once (one copy) instead of
+  re-concatenating the backlog per feed.
+
+## What Stays Fast / Unchanged
+
+- Relay PTY output batching (interactive echo fast path in
+  `src/relay/pty-handler.ts`) is untouched.
+- No `listSessions()` / provider inventory was added anywhere near typing,
+  focus, switch, resume, resize, or render paths.
+- Remote file streaming correctness: per-chunk length checks, chunk-count and
+  byte-count invariants, cancel paths, and the 12MB round-trip integration
+  test all pass unchanged.
+
+## Regression Coverage
+
+- `src/relay/fs-stream-pty-echo-backpressure.integration.test.ts` — real
+  mux ↔ dispatcher ↔ FsHandler over a congestible in-memory pipe; asserts
+  deterministic BYTE bounds (not wall-clock): echo queues behind < 2 framed
+  chunks when congested (pre-fix: whole file), ack window caps in-flight
+  chunks, legacy no-ack clients still get complete streams.
+- `src/relay/fs-handler-stream.test.ts` — pump parks at the ack window,
+  resumes per ack, and releases its file handle when cancelled while parked.
+- `src/relay/dispatcher.test.ts` — notifyBulk semantics: drain gating,
+  per-client targeting, dispose releases parked senders, interactive
+  notify() not gated behind a stalled bulk lane.
+- `src/main/ssh/relay-protocol.test.ts` — decoder byte-at-a-time boundary
+  straddling, oversized-frame resync across odd chunk sizes, and a
+  no-`Buffer.concat`-during-feed guard that locks out the O(n²) shape.
+- `tests/e2e/ssh-docker-relay-perf.spec.ts` — new "busy relay" scenario:
+  types while two 8MB remote file-read loops and a git.status loop run;
+  budgets median < 500ms, worst < 2000ms. (On loopback Docker the pre-fix
+  HOL is only tens of ms — the in-process byte-bound test is the
+  authoritative red/green; the e2e guards the end-to-end wiring.)
+
+## Residual Gaps
+
+- **Large single-frame responses**: `git.*` responses (stdout capped at
+  `MAX_GIT_BUFFER` = 10MB → ~13MB framed) and other request/response results
+  are one atomic frame on the wire; a huge diff can still delay an echo by
+  its own transfer time. Fixing this requires response chunking (a protocol
+  change); bounded by the 16MB `MAX_MESSAGE_SIZE`.
+- **Outbound direction**: a large `fs.writeFile` request frame (client →
+  relay) can queue ahead of keystroke `pty.data` frames on `channel.stdin`.
+  Same shape, opposite direction, rarer trigger (saving a large remote file
+  while typing).
+- `pty.ackData` flow control for PTY output remains unenforced
+  (`src/relay/pty-handler.ts` — "not yet enforced"); PTY output floods are
+  already paced by the relay's 8ms batch/16KB slice scheduler.

docs/source-control-push-failure-ai-recovery.md

@@ -0,0 +1,368 @@
+# Source Control Push Failure AI Recovery
+
+## Problem
+
+Push, force-push, publish, and sync push-stage failures flow through the shared
+remote operation formatter in `src/renderer/src/lib/source-control-remote-error.ts:94`
+and the editor-store toast paths in `src/renderer/src/store/slices/editor.ts:3716`
+and `src/renderer/src/store/slices/editor.ts:3821`. A local pre-push hook or
+lint hook is not an auth, protected-branch, or transport problem, so generic
+remote guidance sends the user in the wrong direction.
+
+PR #7787 proves the behavior is useful, but the prototype needs tightening before
+it is maintainable:
+
+- `src/renderer/src/components/right-sidebar/SourceControl.tsx:1938` and
+  `src/renderer/src/components/right-sidebar/SourceControl.tsx:5799` repeat
+  push-failure predicates and feed live branch/file state into a prompt that says
+  "at failure time".
+- `src/renderer/src/components/right-sidebar/use-source-control-recovery-ai.ts:63`
+  imports push detection again instead of consuming one derived recovery model.
+- `src/renderer/src/components/right-sidebar/source-control-ai-commit-failure-launch.ts:24`
+  and `src/renderer/src/components/right-sidebar/source-control-ai-push-failure-launch.ts:24`
+  are copied launch flows with action-specific copy.
+- `src/shared/source-control-commit-failure-agent-command.ts:6` and
+  `src/shared/source-control-push-failure-agent-command.ts:6` are identical
+  command-template builders.
+- `src/renderer/src/components/right-sidebar/source-control-fix-split-button.tsx:21`
+  already contains a reusable split button, while `SourceControl.tsx:6415`
+  carries a second inline version.
+- `src/renderer/src/lib/source-control-remote-error.ts:148` currently treats
+  any `isSync` error as push-like for hook detection. `syncBranch` calls the
+  formatter for fetch, upstream-status, pull, and push failures, so the renderer
+  cannot prove a sync error came from the push stage without an explicit marker.
+- `src/renderer/src/components/right-sidebar/SourceControl.push-failure-recovery.test.ts:2`
+  imports prompt helpers through the large React module instead of shared prompt
+  modules.
+
+## Goal
+
+Add first-class recovery for pre-push hook failures without growing a parallel
+push-only subsystem:
+
+1. Toasts for push, force-push, publish, and sync push-stage failures say
+   "blocked" for detected pre-push or lint-hook output.
+2. Source Control shows a concise Push blocked panel with Details and AI Fix
+   only for a failure snapshot that is both push-like and hook-like.
+3. AI Fix launches the configured `fixPushFailure` Source Control action with a
+   safe, provider-neutral prompt.
+4. Commit-failure recovery and push-failure recovery share launch,
+   command-template, split-button, and dense error-panel primitives.
+
+## Non-goals
+
+- Do not fix server-side pre-receive hooks, hosted CI failures, or provider-side
+  protected-branch failures from this entry point.
+- Do not bypass hooks, add `--no-verify`, push from the launched agent, create a
+  PR, or assume GitHub-specific terminology.
+- Do not add app-wide persistence for this transient failure panel.
+- Do not redesign the Source Control panel or the Source Control AI settings
+  model.
+- Do not fold unrelated launch surfaces such as checks recovery into this change
+  unless a tiny shared helper is already needed for commit/push recovery.
+
+## Design
+
+1. Keep push-hook detection and prompt building in shared code.
+   - `src/shared/source-control-push-failure.ts` owns normalization, bounded
+     scanning, summary text, Details eligibility, prompt truncation, prompt
+     rules, and changed-file list bounding.
+   - Detection must stay conservative: explicit `pre-push`/`prepush`,
+     `hook declined to push`, hook-runner output in push context, or lint output
+     in push context. Auth, non-fast-forward, protected branch, pre-receive,
+     submodule, and generic transport failures stay on the existing remote-error
+     path.
+   - Keep the output scan bounded at 64 KiB and the prompt failure-output
+     section bounded. Also cap prompt file lines to a small constant with an
+     omitted-count line; do not rely on the git-status cap, which is far too high
+     for prompt context.
+
+2. Make remote-error formatting sync-stage aware.
+   - Extend `RemoteOperationErrorOptions` with an explicit push-stage flag for
+     sync, for example `isSyncPushStage`.
+   - Gate push-hook classification on `isPush`, `isForcePush`, `publish`, or
+     `isSyncPushStage`; do not use bare `isSync` as evidence of a push failure.
+   - In `syncBranch`, pass the push-stage flag only inside the two inner
+     `pushRuntimeGit` catch blocks. The outer fetch, upstream-status, and pull
+     catches still pass `isSync` for Sync-shaped generic copy but must not render
+     blocked hook copy.
+   - Carry the same push-stage fact back to the renderer, either by tagging the
+     thrown `Error` with a narrow exported marker or by wrapping it in a typed
+     error that preserves the original message/cause. `SourceControl` must set
+     `syncPushStage` only from that marker, not from `kind === 'sync'` plus
+     hook-like stderr.
+   - Keep submodule push messages before hook detection so submodule guidance
+     remains more specific than AI recovery.
+
+3. Replace one-off command builders with one shared recovery command builder.
+   - Add a shared builder that accepts `{ actionId, promptOverride,
+     commandInputTemplate, basePrompt }` and renders the existing Source Control
+     AI command template for launch actions.
+   - Keep compatibility exports for commit and push helpers if nearby tests or
+     call sites still use those names, but make them wrappers over the generic
+     builder.
+   - Move prompt-builder tests to import from shared modules, not
+     `SourceControl.tsx`.
+
+4. Capture and derive push recovery state once.
+   - Extend `SourceControlActionError` with the raw error, a push-stage marker
+     for sync failures, the branch name at failure time, a bounded status-entry
+     snapshot at failure time, and a per-worktree sequence token.
+   - `runRemoteAction` clears the worktree's previous error and increments its
+     sequence before starting. A catch may write an error only if its sequence
+     still owns that worktree, so a slow failure cannot overwrite a newer retry
+     or success.
+   - Add a focused renderer helper that accepts the captured
+     `SourceControlActionError` and the current branch name, then returns either
+     `null` or a model with raw/sanitized detail text, summary, details flag,
+     kind label, and AI prompt.
+   - The helper returns `null` unless the operation is `push`, `force_push`,
+     `publish`, or `sync` with the sync push-stage marker and
+     `isPushHookFailure(rawError)` passes.
+   - If the current branch is known and differs from the captured branch, hide or
+     clear the model instead of launching an agent with wrong branch context.
+   - Pass that single model to `useSourceControlAi` and `CommitArea`; neither
+     prop construction nor `CommitArea` should repeat the push-hook predicate.
+
+5. Use one generic recovery launcher for commit and push failures.
+   - Replace the separate commit and push launch files with
+     `source-control-ai-recovery-launch.ts`, or keep the old files as one-line
+     wrappers if import stability is cheaper.
+   - The generic launcher owns connection resolution, SSH/local agent discovery,
+     saved-agent validation, agent-args validation, agent selection, terminal
+     launch, focus, and success/failure toasts.
+   - Resolve connection without collapsing local and unresolved states: read
+     `const worktreeConnectionId = getConnectionId(worktreeId)`, use it when it
+     is a string or `null`, and only fall back to `sourceRepoConnectionId` when
+     the worktree lookup returns `undefined`. If the fallback is also
+     `undefined`, show the workspace-connection error instead of launching
+     locally. `null` means proven local; a string means SSH/remote.
+   - Validate saved CLI args before agent detection or terminal creation. On
+     Windows, keep using PowerShell-safe planning through
+     `planAgentCliArgsSuffix(..., 'powershell')`.
+   - Action-specific data is limited to action id, base prompt, empty-prompt
+     copy, unavailable-agent copy, and success copy.
+
+6. Keep the recovery hook thin.
+   - `use-source-control-recovery-ai.ts` builds the commit prompt, accepts the
+     already-derived push recovery prompt/model, keeps independent loading
+     flags, and calls the generic launcher.
+   - It should not import push-hook detection, resolve agents, or duplicate
+     launch plumbing.
+
+7. Extract and reuse the recovery UI.
+   - Reuse or replace the existing
+     `src/renderer/src/components/right-sidebar/source-control-fix-split-button.tsx`
+     rather than adding another split-button component.
+   - Move the dense recovery notice and Details dialog into focused renderer
+     modules named for source-control recovery. Avoid `helpers`, `utils`, and new
+     max-lines suppressions.
+   - Reuse the same notice/dialog component for commit and push recovery with
+     action-specific labels, summary, details, prompt, saved recipe, and launch
+     callback.
+   - Keep normal remote errors filtered out when a push recovery model is
+     rendered, so the user does not see both Push blocked and generic remote
+     error copy for the same failure.
+
+8. Retain `fixPushFailure` in the existing Source Control AI action model.
+   - The prototype already adds `fixPushFailure` to
+     `src/shared/source-control-ai-actions.ts`; keep it as a launch action like
+     `fixCommitFailure`.
+   - Verify default settings, normalizers, global settings rows, repository
+     override rows, labels, descriptions, and variable chips all include Push
+     failure fixes.
+   - Do not create a push-specific settings store or migration. Existing
+     `normalizeSourceControlAiSettings` hydrates missing action defaults.
+
+## Data Flow
+
+- User clicks Push, Force Push, Publish Branch, or Sync.
+- `runRemoteAction` clears that worktree's previous remote error, records a new
+  sequence token, and starts the editor-store remote action.
+- Git push fails and throws.
+- `pushBranch` or sync's inner push-stage catch calls
+  `resolveRemoteOperationErrorMessage` with push-like options. Sync fetch, pull,
+  and upstream-status failures do not pass the push-stage flag.
+- Sync's inner push-stage catch marks the rethrown error as push-stage so
+  `runRemoteAction` can preserve `syncPushStage` in the captured snapshot.
+- The formatter classifies hook output before auth/transport fallbacks only for
+  push-like options and shows blocked toast copy.
+- `SourceControl` catches the same error, and if the sequence still owns the
+  worktree, stores `{ kind, rawError, message, syncPushStage, branchName,
+  entriesSnapshot }`.
+- The push recovery helper derives one `pushRecovery` model from that snapshot.
+- `CommitArea` renders normal remote errors or the Push blocked recovery notice.
+- AI Fix builds the `fixPushFailure` command input and launches the selected
+  agent in the owning local or SSH runtime.
+
+## Edge Cases
+
+- Auth, missing repo, protected-branch, pre-receive, non-fast-forward, and
+  transport failures must not show the Push blocked panel.
+- Submodule push errors keep their existing specialized messages before
+  push-hook detection.
+- Sync failures from fetch, upstream-status, or pull stages must not show Push
+  blocked, even if stderr contains words like "lint" or "hook".
+- Create PR intent remote failures use the same `runRemoteAction` plumbing; if
+  CommitArea is not visible, blocked toast copy is still required but no hidden
+  panel work is needed.
+- A newer remote operation for the same worktree must clear stale recovery state
+  immediately and prevent an older in-flight failure from writing after it.
+- Switching worktrees or branches must not show another worktree or branch's
+  push failure.
+- External edits or another Orca window may make the failure stale. Do not add
+  cross-window persistence for this feature; rely on local retry/remount/branch
+  mismatch clearing and list this as residual risk.
+- Source Control AI hidden for a repo still shows the blocked notice and Details;
+  it hides AI Fix.
+- SSH worktrees must detect and launch agents on the owning connection, not a
+  local fallback.
+- Windows launch planning must continue to use PowerShell-safe agent args.
+- Very large hook output and very large changed-file sets must be bounded before
+  prompt generation.
+- ANSI/control output must not leak into summaries or Details comparison logic.
+  Details may show sanitized text while preserving enough line breaks for
+  debugging.
+- Empty custom command templates must stay empty so the launcher rejects them
+  with a clear settings error.
+- Prompt text must treat file paths, branch names, and hook output as data, not
+  instructions.
+
+## Test Plan
+
+- Shared unit tests:
+  - `src/shared/source-control-push-failure.test.ts` covers detection positives,
+    auth/protected/pre-receive/non-fast-forward negatives, ANSI/control
+    stripping, bounded scanning, details comparison, prompt output truncation,
+    prompt file-list capping, and provider-neutral prompt rules.
+  - Shared command-template tests cover the generic recovery builder plus
+    commit/push wrappers, including empty templates and prompt overrides.
+  - `src/shared/source-control-ai-actions.test.ts` covers `fixPushFailure`
+    label, default template, variables, normalization, and action-list inclusion.
+- Remote formatter and store tests:
+  - `src/renderer/src/lib/source-control-remote-error.test.ts` covers push,
+    force-push, publish, and sync push-stage blocked copy plus sync non-push
+    stage negatives, auth/non-fast-forward/protected/pre-receive/submodule
+    regressions.
+  - `src/renderer/src/store/slices/editor.test.ts` covers the `pushBranch` toast
+    path and both sync push-stage and sync non-push-stage toasts.
+- Renderer model and UI tests:
+  - A focused push recovery derivation test covers operation kind, sync
+    push-stage marker, branch mismatch, stale sequence ownership, snapshot file
+    entries, summary, details flag, prompt contents, and ordinary remote errors.
+  - `CommitArea` tests cover the Push blocked notice, Details dialog, AI Fix
+    visible/hidden, ordinary remote errors, no duplicate remote error when a push
+    model exists, and unchanged commit-failure recovery rendering.
+  - Settings tests cover global and repository Source Control AI action rows
+    showing Push failure fixes.
+- Launcher tests:
+  - Generic recovery launcher tests cover commit and push action ids, invalid CLI
+    args before detection, empty template rejection, unavailable saved agent,
+    local versus SSH agent detection, successful launch/focus, and success copy.
+- Checks:
+  - Run targeted vitest files above, `pnpm typecheck`, `pnpm lint`, and
+    `pnpm check:max-lines-ratchet`.
+
+## UI Quality Bar
+
+The Source Control recovery UI must match `docs/STYLEGUIDE.md`: monochrome,
+quiet, dense, token-based, and consistent with adjacent commit-failure recovery.
+Use existing shadcn `Button`, `DropdownMenu`, and `Dialog` primitives, lucide
+icons, `card`/`border`/`destructive`/`muted` tokens, and the existing action
+recipe row patterns. The Push blocked notice should keep the summary readable at
+narrow right-sidebar widths, avoid nested cards, avoid new color values, show
+stable split-button geometry while launching, and keep Details as progressive
+disclosure for long hook output. Dialogs and dropdowns must remain usable in
+light/dark mode and under SSH latency.
+
+## Review Screenshots
+
+1. Source Control after a simulated pre-push hook failure: Push blocked notice
+   with summary, AI Fix, and Details.
+2. Push failure Details dialog showing hook output and Fix with AI.
+3. AI Fix customize-launch dialog for `fixPushFailure`.
+4. Ordinary push auth/protected/remote failure state: regular remote error, no
+   Push blocked notice.
+5. Sync fetch/pull-stage failure state: Sync-shaped remote error, no Push
+   blocked notice.
+6. Source Control AI hidden for the repo: Push blocked notice and Details, no AI
+   Fix.
+7. Source Control AI settings showing the Push failure fixes launch recipe.
+8. Adjacent commit-failure recovery notice still rendering correctly after the
+   shared UI extraction.
+
+## Rollout
+
+1. Tighten shared push-failure classifier/prompt bounds and add the generic
+   recovery command builder.
+2. Make remote formatter/store sync-stage aware and update toast tests.
+3. Add the `SourceControlActionError` snapshot/sequence data and one push
+   recovery derivation helper.
+4. Replace duplicated commit/push launchers with the generic recovery launcher
+   and thin hook wiring.
+5. Extract/reuse source-control recovery UI components, including the existing
+   split-button module.
+6. Verify `fixPushFailure` settings/default/normalizer coverage and i18n labels.
+7. Update targeted tests and run typecheck, lint, and max-lines ratchet.
+8. Validate rendered Electron states with screenshots before opening the PR.
+
+## Lightweight Eng Review
+
+- Scope: Reduced from PR #7787's push-specific parallel path to a shared
+  recovery layer. Keep the feature limited to local pre-push/lint-hook detection,
+  blocked copy, an inline recovery notice, Details, and AI launch. No persistence
+  or provider-specific recovery.
+- Architecture/data flow: Shared code owns classification, prompt text, and
+  command-template rendering; the editor store owns stage-aware toast formatting;
+  `SourceControl` owns transient worktree-scoped failure snapshots and sequence
+  invalidation; the generic launcher owns local/SSH agent discovery and terminal
+  launch.
+- Failure modes covered:
+  - Sync stage ambiguity: only the inner push stage may classify as hook-blocked.
+  - Auth/protected/pre-receive/non-fast-forward/submodule/transport errors stay
+    out of push recovery.
+  - Branch or worktree switches suppress stale recovery models.
+  - Per-worktree sequence tokens prevent older in-flight failures from
+    overwriting newer retries or successes.
+  - Empty custom templates, invalid CLI args, unavailable saved agents, missing
+    worktree context, SSH host detection, and Windows shell planning fail before
+    terminal launch.
+  - Large or hostile hook output and file lists are bounded and treated as data
+    in prompts.
+- Test coverage required:
+  - Shared classifier/prompt/command/action-model tests in `src/shared`.
+  - Remote formatter and editor-store tests for push, force-push, publish, sync
+    push-stage, and sync non-push-stage paths.
+  - Renderer derivation and `CommitArea` tests for Push blocked, Details, AI Fix,
+    hidden AI actions, stale branch/sequence, ordinary remote errors, and
+    existing commit failure recovery.
+  - Generic launcher tests for local/SSH, invalid args, saved-agent unavailable,
+    empty template, and success launch.
+  - Settings tests for global and repo Push failure fixes rows.
+  - Electron validation for all required screenshots; no separate E2E is
+    required if renderer tests cover the model and Electron validates the real UI.
+- Performance/blast radius: No polling, watchers, migrations, or startup work.
+  Runtime cost is one bounded classifier pass over at most 64 KiB, bounded prompt
+  output, and a bounded file snapshot only when a remote action fails. Largest
+  blast radius is the shared recovery UI and launcher used by existing commit
+  failures, so commit recovery needs tests and a screenshot.
+- UI quality bar: Match `docs/STYLEGUIDE.md` and adjacent Source Control density:
+  token-based error state, stable split-button geometry, progressive Details
+  disclosure, no nested cards, concise copy, light/dark compatibility, and
+  visible disabled/loading state under SSH latency.
+- Required review screenshots:
+  1. Push blocked notice after pre-push hook failure.
+  2. Push failure Details dialog with Fix with AI.
+  3. `fixPushFailure` customize-launch dialog.
+  4. Ordinary push auth/protected/remote failure with no Push blocked panel.
+  5. Sync non-push-stage failure with no Push blocked panel.
+  6. Source Control AI hidden: Push blocked without AI Fix.
+  7. Source Control AI settings showing Push failure fixes.
+  8. Existing commit-failure recovery notice after shared UI extraction.
+- Residual risks: Classifier false positives remain possible for unusual local
+  hook-runner output that mentions push context; keep tests biased toward
+  preserving auth, protected-branch, and non-fast-forward behavior. Failure
+  panels are renderer-local, so another Orca window or an external terminal can
+  fix the problem while the old panel remains visible until local retry, remount,
+  or branch/worktree change. Electron validation may need a mocked or temporary
+  repo scenario to trigger a real pre-push hook without mutating user data.

electron.vite.config.ts

@@ -2,6 +2,7 @@ import { resolve } from 'node:path'
 import { defineConfig } from 'electron-vite'
 import react from '@vitejs/plugin-react'
 import tailwindcss from '@tailwindcss/vite'
+import { createPlainNodeEntryGuardPlugin } from './build-plugins/plain-node-entry-guard'
 
 // Why: the telemetry transport is gated by two compile-time constants that
 // only the official CI release workflow sets. Contributor / `pnpm dev` /
@@ -182,13 +183,16 @@ export default defineConfig({
           'stt-worker': resolve('src/main/speech/stt-worker.ts'),
           'warp-theme-parser-worker': resolve('src/main/warp-themes/warp-theme-parser-worker.ts'),
           'file-watcher-worker': resolve('src/main/runtime/file-watcher-worker.ts'),
+          // Why: forked with ELECTRON_RUN_AS_NODE so @parcel/watcher faults
+          // can't take down the main process (issue #7547).
+          'parcel-watcher-process-entry': resolve('src/main/ipc/parcel-watcher-process-entry.ts'),
           // Why: electron-vite cleans out/main in dev. The dev CLI imports
           // this path for `orca agent hooks ...`, so it must survive rebuilds.
           'agent-hooks/managed-agent-hook-controls': resolve(
             'src/main/agent-hooks/managed-agent-hook-controls.ts'
           )
         },
-        plugins: [createStartupDiagnosticsBootstrapPlugin()]
+        plugins: [createStartupDiagnosticsBootstrapPlugin(), createPlainNodeEntryGuardPlugin()]
       }
     },
     // Why: compile-time substitution for the telemetry gate. See the block

mobile/.gitignore

@@ -1,4 +1,5 @@
 node_modules/
+src/terminal/terminal-webview-engine.generated.ts
 .expo/
 dist/
 /android/

mobile/.oxlintrc.json

@@ -1,6 +1,7 @@
 {
   "$schema": "./node_modules/oxlint/configuration_schema.json",
   "extends": ["../.oxlintrc.json"],
+  "ignorePatterns": ["src/terminal/terminal-webview-engine.generated.ts"],
   "rules": {
     "react-hooks/exhaustive-deps": "off",
     "react/no-unescaped-entities": "off",
@@ -35,7 +36,7 @@
     {
       "files": ["src/terminal/terminal-webview-html.ts"],
       "rules": {
-        "max-lines": ["error", { "max": 1778, "skipBlankLines": true, "skipComments": true }]
+        "max-lines": ["error", { "max": 1784, "skipBlankLines": true, "skipComments": true }]
       }
     },
     {

mobile/app.json

@@ -2,7 +2,7 @@
   "expo": {
     "name": "Orca",
     "slug": "orca-mobile",
-    "version": "0.0.22",
+    "version": "0.0.25",
     "orientation": "default",
     "icon": "./assets/icon.png",
     "userInterfaceStyle": "automatic",
@@ -75,7 +75,7 @@
       "allowBackup": false,
       "permissions": ["RECORD_AUDIO", "MODIFY_AUDIO_SETTINGS"],
       "package": "com.stably.orca.mobile",
-      "versionCode": 2
+      "versionCode": 6
     },
     "plugins": [
       "expo-router",

mobile/app/_layout.tsx

@@ -102,6 +102,14 @@ export default function RootLayout() {
         return
       }
       handledNotificationIdsRef.current.add(notificationId)
+      // Why: RootLayout never unmounts, so cap this tap-dedup set (FIFO) rather
+      // than letting it grow one id per notification tapped for the app's life.
+      if (handledNotificationIdsRef.current.size > 256) {
+        const oldest = handledNotificationIdsRef.current.values().next().value
+        if (oldest !== undefined) {
+          handledNotificationIdsRef.current.delete(oldest)
+        }
+      }
 
       const path = await getNavigationPath(response.notification.request.content.data)
       clearLastNotificationResponse()

mobile/app/h/[hostId]/history/[worktreeId].tsx

@@ -3,14 +3,15 @@ import { ActivityIndicator, FlatList, Pressable, StyleSheet, Text, View } from '
 import { SafeAreaView, useSafeAreaInsets } from 'react-native-safe-area-context'
 import { useLocalSearchParams, useRouter } from 'expo-router'
 import { ChevronDown, ChevronLeft, ChevronRight } from 'lucide-react-native'
-import { useHostClient } from '../../../../src/transport/client-context'
+import { useForceReconnect, useHostClient } from '../../../../src/transport/client-context'
 import type { RpcSuccess } from '../../../../src/transport/types'
-import { colors, spacing, typography } from '../../../../src/theme/mobile-theme'
+import { colors, radii, spacing, typography } from '../../../../src/theme/mobile-theme'
 import {
   fetchMobileGitHistory,
   mapMobileCommitRows,
   type MobileCommitRow
 } from '../../../../src/source-control/mobile-git-history'
+import { resolveMobileHistoryScreenView } from '../../../../src/source-control/mobile-history-screen-state'
 import type { GitBranchChangeEntry } from '../../../../../src/shared/types'
 
 function firstParam(value: string | string[] | undefined): string {
@@ -27,9 +28,11 @@ export default function HistoryScreen() {
   const router = useRouter()
   const insets = useSafeAreaInsets()
   const { client, state: connState } = useHostClient(hostId)
+  const forceReconnect = useForceReconnect()
 
   const [rows, setRows] = useState<MobileCommitRow[] | null>(null)
   const [error, setError] = useState<string | null>(null)
+  const [reloadNonce, setReloadNonce] = useState(0)
   const [expanded, setExpanded] = useState<string | null>(null)
   const [filesById, setFilesById] = useState<Record<string, GitBranchChangeEntry[] | 'loading'>>({})
 
@@ -57,7 +60,20 @@ export default function HistoryScreen() {
     return () => {
       active = false
     }
-  }, [client, connState, worktreeId])
+  }, [client, connState, reloadNonce, worktreeId])
+
+  const retry = useCallback(() => {
+    setError(null)
+    // Why: retrying the fetch is useless while the transport's reconnect loop
+    // is parked at its backoff cap — revive the connection instead (mirrors
+    // MobileSourceControlPanel / issue #5049). The load effect re-runs via
+    // connState once the fresh client connects.
+    if (connState !== 'connected' && hostId) {
+      void forceReconnect(hostId)
+      return
+    }
+    setReloadNonce((n) => n + 1)
+  }, [connState, forceReconnect, hostId])
 
   const toggleCommit = useCallback(
     (row: MobileCommitRow) => {
@@ -130,6 +146,12 @@ export default function HistoryScreen() {
     [expanded, filesById, toggleCommit]
   )
 
+  const view = resolveMobileHistoryScreenView({
+    connected: client !== null && connState === 'connected',
+    rows,
+    error
+  })
+
   return (
     <SafeAreaView style={styles.container} edges={['top']}>
       <View style={styles.header}>
@@ -138,21 +160,26 @@ export default function HistoryScreen() {
         </Pressable>
         <Text style={styles.title}>Commit History</Text>
       </View>
-      {error ? (
+      {view.kind === 'error' || view.kind === 'waiting' ? (
         <View style={styles.state}>
-          <Text style={styles.stateText}>{error}</Text>
+          <Text style={styles.stateText}>
+            {view.kind === 'waiting' ? 'Waiting for desktop...' : view.message}
+          </Text>
+          <Pressable style={styles.retryButton} onPress={retry} accessibilityLabel="Retry">
+            <Text style={styles.retryText}>Retry</Text>
+          </Pressable>
         </View>
-      ) : rows === null ? (
+      ) : view.kind === 'loading' ? (
         <View style={styles.state}>
           <ActivityIndicator color={colors.textSecondary} />
         </View>
-      ) : rows.length === 0 ? (
+      ) : view.kind === 'empty' ? (
         <View style={styles.state}>
           <Text style={styles.stateText}>No commits.</Text>
         </View>
       ) : (
         <FlatList
-          data={rows}
+          data={view.rows}
           renderItem={renderCommit}
           keyExtractor={(row) => row.id}
           contentContainerStyle={{ paddingBottom: spacing.lg + insets.bottom }}
@@ -175,6 +202,14 @@ const styles = StyleSheet.create({
   title: { color: colors.textPrimary, fontSize: typography.bodySize, fontWeight: '600' },
   state: { flex: 1, alignItems: 'center', justifyContent: 'center', padding: spacing.lg },
   stateText: { color: colors.textMuted, fontSize: typography.bodySize },
+  retryButton: {
+    marginTop: spacing.md,
+    paddingHorizontal: spacing.lg,
+    paddingVertical: spacing.sm,
+    borderRadius: radii.button,
+    backgroundColor: colors.bgRaised
+  },
+  retryText: { color: colors.textPrimary, fontSize: typography.bodySize, fontWeight: '600' },
   commit: { borderBottomWidth: 1, borderBottomColor: colors.borderSubtle },
   commitHeader: {
     flexDirection: 'row',

mobile/app/h/[hostId]/index.tsx

@@ -158,7 +158,6 @@ export function HostScreen({
   const now = useNow(30_000)
   const [repoColorsByName, setRepoColorsByName] = useState<Map<string, string>>(new Map())
   const [repoIconsByName, setRepoIconsByName] = useState<Map<string, RepoIcon>>(new Map())
-  const [repoSummaries, setRepoSummaries] = useState<RepoSummary[]>([])
   const [hostName, setHostName] = useState('')
   const [error, setError] = useState('')
   const [compatVerdict, setCompatVerdict] = useState<CompatVerdict>({ kind: 'ok' })
@@ -205,9 +204,7 @@ export function HostScreen({
     hideDefaultBranch: false,
     filterRepoIds: [],
     collapsedGroups: [],
-    workspaceStatuses: DEFAULT_MOBILE_WORKSPACE_STATUSES,
-    workspaceHostScope: undefined,
-    visibleWorkspaceHostIds: undefined
+    workspaceStatuses: DEFAULT_MOBILE_WORKSPACE_STATUSES
   })
 
   useEffect(() => {
@@ -218,9 +215,7 @@ export function HostScreen({
       hideDefaultBranch: filters.hideDefaultBranch,
       filterRepoIds: [...filters.filterRepoIds],
       collapsedGroups: [...collapsedGroups],
-      workspaceStatuses,
-      workspaceHostScope: viewStateRef.current.workspaceHostScope,
-      visibleWorkspaceHostIds: viewStateRef.current.visibleWorkspaceHostIds
+      workspaceStatuses
     }
   }, [groupMode, sortMode, filters, collapsedGroups, workspaceStatuses])
 
@@ -339,7 +334,6 @@ export function HostScreen({
     setCompatVerdict({ kind: 'ok' })
     setRepoColorsByName(new Map())
     setRepoIconsByName(new Map())
-    setRepoSummaries([])
     repoMetadataFetchedAtRef.current = 0
     // Why: re-seed from the current host's cache on every hostId change.
     // The useState initializer only runs on first mount, so if Expo Router
@@ -398,7 +392,6 @@ export function HostScreen({
         const repoResult = (repoResponse as RpcSuccess).result as { repos: RepoSummary[] }
         repoMetadataFetchedAtRef.current = Date.now()
         setCachedRepos(requestHostId, repoResult.repos)
-        setRepoSummaries(repoResult.repos)
         setRepoColorsByName(
           new Map(
             repoResult.repos.map((repo) => [
@@ -821,11 +814,8 @@ export function HostScreen({
     groupMode,
     pinnedIds,
     repoIdsByName,
-    repoSummaries,
     repoColorsByName,
     collapsedGroups,
-    workspaceHostScope: viewStateRef.current.workspaceHostScope,
-    visibleWorkspaceHostIds: viewStateRef.current.visibleWorkspaceHostIds,
     workspaceStatuses
   })
   const existingWorktreePaths = useMemo(() => worktrees.map((w) => w.path), [worktrees])

mobile/app/h/[hostId]/session/[worktreeId].tsx

@@ -24,6 +24,7 @@ import {
   AlertTriangle,
   ArrowUp,
   Bot,
+  ChevronDown,
   ChevronLeft,
   ChevronRight,
   ChevronsRight,
@@ -96,9 +97,11 @@ import { createTerminalLiveAccessoryInput } from '../../../../src/terminal/termi
 import { getTerminalLiveAccessoryRawSendTarget } from '../../../../src/terminal/terminal-live-accessory-raw-send-target'
 import {
   clearTerminalLiveInputFocusTimer,
+  focusTerminalLiveInputTarget,
   isTerminalLiveInputWithinByteLimit,
   scheduleTerminalLiveInputFocus
 } from '../../../../src/terminal/terminal-live-input'
+import { dismissTerminalKeyboard } from '../../../../src/terminal/terminal-keyboard-dismiss'
 import type { TerminalLiveInputSender } from '../../../../src/terminal/terminal-live-input-sender'
 import { isTerminalSendRpcAccepted } from '../../../../src/terminal/terminal-send-rpc-response'
 import { useTerminalLiveInputCommit } from '../../../../src/terminal/use-terminal-live-input-commit'
@@ -1005,6 +1008,7 @@ export default function SessionScreen() {
   const viewportMeasuredRef = useRef(false)
   const terminalRefs = useRef<Map<string, TerminalWebViewHandle>>(new Map())
   const liveInputRef = useRef<TextInput>(null)
+  const commandInputRef = useRef<TextInput>(null)
   const liveInputFocusTimerRef = useRef<ReturnType<typeof setTimeout> | null>(null)
   const sendLiveTerminalInputRef = useRef<TerminalLiveInputSender>(async () => false)
   const sessionTabActionSheetKeyboardHideSubRef = useRef<ReturnType<
@@ -3134,8 +3138,12 @@ export default function SessionScreen() {
     if (!canSend || !liveInputEnabled) {
       return
     }
-    liveInputRef.current?.focus()
-  }, [canSend, liveInputEnabled])
+    focusTerminalLiveInputTarget(liveInputRef.current, {
+      keyboardHeight,
+      refocus: () =>
+        scheduleTerminalLiveInputFocus(liveInputFocusTimerRef, () => liveInputRef.current?.focus())
+    })
+  }, [canSend, keyboardHeight, liveInputEnabled])
 
   const clearSessionTabActionSheetKeyboardListener = useCallback(() => {
     sessionTabActionSheetKeyboardHideSubRef.current?.remove()
@@ -3203,6 +3211,15 @@ export default function SessionScreen() {
     ]
   )
 
+  const dismissSoftwareKeyboard = useCallback(() => {
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus: () => clearTerminalLiveInputFocusTimer(liveInputFocusTimerRef),
+      commandInput: commandInputRef.current,
+      dismissKeyboard: () => Keyboard.dismiss(),
+      liveInput: liveInputRef.current
+    })
+  }, [])
+
   const handleTerminalTap = useCallback(
     (handle: string) => {
       if (handle !== activeHandleRef.current) {
@@ -4566,28 +4583,27 @@ export default function SessionScreen() {
                     </View>
                   </Pressable>
                 ))}
-                <Pressable
-                  style={({ pressed }) => [
-                    styles.newTerminalButton,
-                    pressed && styles.newTerminalButtonPressed,
-                    (creating ||
-                      creatingBrowser ||
-                      creatingMarkdown ||
-                      connState !== 'connected') &&
-                      styles.newTerminalButtonDisabled
-                  ]}
-                  disabled={
-                    creating || creatingBrowser || creatingMarkdown || connState !== 'connected'
-                  }
-                  onPress={() => {
-                    setCreateError('')
-                    setShowCreateTabDrawer(true)
-                  }}
-                  accessibilityLabel="New tab"
-                >
-                  <Plus size={16} color={colors.textSecondary} strokeWidth={2.2} />
-                </Pressable>
               </ScrollView>
+              {/* Why: pinned outside the scroll strip so the new-agent button
+                  stays reachable no matter how far the tabs scroll. */}
+              <Pressable
+                style={({ pressed }) => [
+                  styles.newTerminalButton,
+                  pressed && styles.newTerminalButtonPressed,
+                  (creating || creatingBrowser || creatingMarkdown || connState !== 'connected') &&
+                    styles.newTerminalButtonDisabled
+                ]}
+                disabled={
+                  creating || creatingBrowser || creatingMarkdown || connState !== 'connected'
+                }
+                onPress={() => {
+                  setCreateError('')
+                  setShowCreateTabDrawer(true)
+                }}
+                accessibilityLabel="New tab"
+              >
+                <Plus size={16} color={colors.textSecondary} strokeWidth={2.2} />
+              </Pressable>
             </View>
           )}
         </SafeAreaView>
@@ -4776,10 +4792,38 @@ export default function SessionScreen() {
               >
                 {/* Accessory keys */}
                 <View style={styles.accessoryBar}>
+                  {/* Why: a fixed, always-visible escape hatch from the open
+                  keyboard. Kept outside the horizontal ScrollView so it does
+                  not scroll away, and out of the terminal-byte shortcut path so
+                  it cannot be hidden by user shortcut customization (#5106). */}
+                  {keyboardLift > 0 && (
+                    <Pressable
+                      style={({ pressed }) => [
+                        styles.keyboardDismissKey,
+                        pressed && styles.accessoryKeyPressed
+                      ]}
+                      onPress={dismissSoftwareKeyboard}
+                      hitSlop={8}
+                      accessibilityRole="button"
+                      accessibilityLabel="Dismiss keyboard"
+                      accessibilityHint="Hides the software keyboard and keeps the current terminal session open."
+                    >
+                      <View style={styles.keyboardDismissGlyph}>
+                        <KeyboardIcon size={15} color={colors.textSecondary} strokeWidth={2} />
+                        <ChevronDown
+                          size={10}
+                          color={colors.textSecondary}
+                          strokeWidth={2.5}
+                          style={styles.keyboardDismissChevron}
+                        />
+                      </View>
+                    </Pressable>
+                  )}
                   {/* Why: with default tap handling the first tap on any accessory
                   key dismisses the open keyboard and is swallowed, so live
                   input lost its keyboard on every Esc/Tab press (#5106). */}
                   <ScrollView
+                    style={styles.accessoryScroll}
                     horizontal
                     showsHorizontalScrollIndicator={false}
                     contentContainerStyle={styles.accessoryContent}
@@ -4947,10 +4991,16 @@ export default function SessionScreen() {
                 {liveInputEnabled ? (
                   <View style={[styles.inputBar, styles.liveInputBar]}>
                     <Pressable
-                      style={styles.liveInputFocusTarget}
+                      style={({ pressed }) => [
+                        styles.liveInputFocusTarget,
+                        pressed && styles.liveInputFocusTargetPressed,
+                        !canSend && styles.liveInputFocusTargetDisabled
+                      ]}
                       disabled={!canSend}
                       onPress={focusLiveInput}
-                      accessibilityLabel="Focus live terminal input"
+                      accessibilityRole="button"
+                      accessibilityLabel="Show keyboard for live terminal input"
+                      accessibilityHint="Typed text is sent directly to the active terminal"
                     >
                       <KeyboardIcon size={16} color={colors.textSecondary} strokeWidth={2} />
                       <MobileTerminalLiveInputStatus
@@ -4981,6 +5031,7 @@ export default function SessionScreen() {
                       onKeyPress={handleLiveInputKeyPress}
                       onSubmitEditing={handleLiveInputSubmit}
                       placeholder=""
+                      showSoftInputOnFocus
                       autoCapitalize="none"
                       autoCorrect={false}
                       spellCheck={false}
@@ -4998,6 +5049,7 @@ export default function SessionScreen() {
                 ) : (
                   <View style={styles.inputBar}>
                     <TextInput
+                      ref={commandInputRef}
                       // Why: Android caches the IME inputType at mount, so toggling
                       // autocomplete must remount there; iOS can update without a focus-costly remount.
                       key={
@@ -5009,9 +5061,10 @@ export default function SessionScreen() {
                       }
                       style={styles.textInput}
                       value={input}
-                      onChangeText={(text) =>
-                        setInput((previousText) => normalizeTerminalTextInput(text, previousText))
-                      }
+                      // Why: iOS kills an active dictation/IME session when JS
+                      // writes a value that differs from the native field text;
+                      // store the raw field text and normalize at send time.
+                      onChangeText={setInput}
                       placeholder="Type a command…"
                       placeholderTextColor={colors.textMuted}
                       autoCapitalize="none"

mobile/app/h/[hostId]/session/mobile-session-command-input-styles.ts

@@ -68,10 +68,16 @@ export const mobileSessionCommandInputStyles = StyleSheet.create({
     zIndex: 20
   },
   accessoryBar: {
+    flexDirection: 'row',
+    alignItems: 'center',
     borderTopWidth: 1,
     borderTopColor: colors.borderSubtle,
     backgroundColor: colors.bgPanel
   },
+  accessoryScroll: {
+    flex: 1,
+    minWidth: 0
+  },
   accessoryContent: {
     paddingHorizontal: spacing.sm,
     paddingVertical: spacing.xs,
@@ -110,6 +116,29 @@ export const mobileSessionCommandInputStyles = StyleSheet.create({
   accessoryKeyTextDisabled: {
     color: colors.textMuted
   },
+  keyboardDismissKey: {
+    alignItems: 'center',
+    justifyContent: 'center',
+    marginLeft: spacing.sm,
+    marginVertical: spacing.xs,
+    backgroundColor: colors.bgRaised,
+    paddingHorizontal: spacing.sm + 2,
+    paddingVertical: 0,
+    borderRadius: radii.button,
+    minWidth: 36,
+    height: 28
+  },
+  keyboardDismissGlyph: {
+    alignItems: 'center',
+    height: 18,
+    justifyContent: 'flex-start',
+    position: 'relative',
+    width: 18
+  },
+  keyboardDismissChevron: {
+    bottom: -2,
+    position: 'absolute'
+  },
   inputBar: {
     flexDirection: 'row',
     alignItems: 'center',
@@ -141,7 +170,20 @@ export const mobileSessionCommandInputStyles = StyleSheet.create({
     minHeight: 34,
     flexDirection: 'row',
     alignItems: 'center',
-    gap: spacing.sm
+    gap: spacing.sm,
+    backgroundColor: colors.bgRaised,
+    borderWidth: 1,
+    borderColor: colors.borderSubtle,
+    borderRadius: radii.input,
+    paddingHorizontal: spacing.sm + 2
+  },
+
+  liveInputFocusTargetPressed: {
+    backgroundColor: colors.borderSubtle
+  },
+
+  liveInputFocusTargetDisabled: {
+    opacity: 0.45
   },
 
   liveInputCapture: {

mobile/package.json

@@ -7,6 +7,7 @@
     "start": "node scripts/start-expo.mjs",
     "android": "expo run:android",
     "ios": "expo run:ios",
+    "postinstall": "node scripts/build-terminal-webview-engine.mjs",
     "test": "vitest run",
     "lint": "oxlint",
     "format": "oxfmt --write .",
@@ -18,6 +19,9 @@
   "dependencies": {
     "@orca/expo-two-way-audio": "file:./packages/expo-two-way-audio",
     "@react-native-async-storage/async-storage": "^2.2.0",
+    "@xterm/addon-unicode11": "0.10.0-beta.285",
+    "@xterm/addon-webgl": "0.20.0-beta.284",
+    "@xterm/xterm": "6.1.0-beta.285",
     "buffer": "^6.0.3",
     "expo": "^55.0.23",
     "expo-build-properties": "^55.0.13",
@@ -62,6 +66,8 @@
     "@types/react-native": "^0.73.0",
     "@types/react-test-renderer": "19.1.0",
     "@types/ws": "^8.18.1",
+    "acorn": "8.15.0",
+    "esbuild": "0.25.4",
     "expo-module-scripts": "^55.0.2",
     "oxfmt": "^0.52.0",
     "oxlint": "^1.71.0",

mobile/pnpm-lock.yaml

@@ -14,6 +14,15 @@ importers:
       '@react-native-async-storage/async-storage':
         specifier: ^2.2.0
         version: 2.2.0(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))
+      '@xterm/addon-unicode11':
+        specifier: 0.10.0-beta.285
+        version: 0.10.0-beta.285(@xterm/xterm@6.1.0-beta.285)
+      '@xterm/addon-webgl':
+        specifier: 0.20.0-beta.284
+        version: 0.20.0-beta.284(@xterm/xterm@6.1.0-beta.285)
+      '@xterm/xterm':
+        specifier: 6.1.0-beta.285
+        version: 6.1.0-beta.285
       buffer:
         specifier: ^6.0.3
         version: 6.0.3
@@ -141,9 +150,15 @@ importers:
       '@types/ws':
         specifier: ^8.18.1
         version: 8.18.1
+      acorn:
+        specifier: 8.15.0
+        version: 8.15.0
+      esbuild:
+        specifier: 0.25.4
+        version: 0.25.4
       expo-module-scripts:
         specifier: ^55.0.2
-        version: 55.0.2(@babel/core@7.29.7)(@babel/runtime@7.29.2)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.28.1)(eslint@9.39.4)(expo@55.0.23)(jest@29.7.0(@types/node@25.9.3))(prettier@2.8.8)(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react-refresh@0.14.2)(react-test-renderer@19.2.6(react@19.2.6))(react@19.2.6)
+        version: 55.0.2(@babel/core@7.29.7)(@babel/runtime@7.29.2)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.25.4)(eslint@9.39.4)(expo@55.0.23)(jest@29.7.0(@types/node@25.9.3))(prettier@2.8.8)(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react-refresh@0.14.2)(react-test-renderer@19.2.6(react@19.2.6))(react@19.2.6)
       oxfmt:
         specifier: ^0.52.0
         version: 0.52.0
@@ -161,10 +176,10 @@ importers:
         version: 5.9.3
       vite:
         specifier: ^8.0.16
-        version: 8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
+        version: 8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
       vitest:
         specifier: ^4.1.9
-        version: 4.1.9(@types/node@25.9.3)(jsdom@20.0.3)(vite@8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))
+        version: 4.1.9(@types/node@25.9.3)(jsdom@20.0.3)(vite@8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))
 
 packages:
 
@@ -1157,126 +1172,252 @@ packages:
   '@emnapi/wasi-threads@1.2.2':
     resolution: {integrity: sha512-c95qOXkHdydNKhscBTebqEC1CVAZpyqOfVfBzQ1qgzyl3gfeldUjIggDbIZgDKsHLgnsM+igH7TJ/eAasaVuMA==}
 
+  '@esbuild/aix-ppc64@0.25.4':
+    resolution: {integrity: sha512-1VCICWypeQKhVbE9oW/sJaAmjLxhVqacdkvPLEjwlttjfwENRSClS8EjBz0KzRyFSCPDIkuXW34Je/vk7zdB7Q==}
+    engines: {node: '>=18'}
+    cpu: [ppc64]
+    os: [aix]
+
   '@esbuild/aix-ppc64@0.28.1':
     resolution: {integrity: sha512-Svl7tq8k/08+p6CXPpRjQ1fKX+1odH/BQbb48fV6fj3CWHhsoIOoY87w1oHXm0qEpkIK3ZfVgp0hed3XBXzXMQ==}
     engines: {node: '>=18'}
     cpu: [ppc64]
     os: [aix]
 
+  '@esbuild/android-arm64@0.25.4':
+    resolution: {integrity: sha512-bBy69pgfhMGtCnwpC/x5QhfxAz/cBgQ9enbtwjf6V9lnPI/hMyT9iWpR1arm0l3kttTr4L0KSLpKmLp/ilKS9A==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [android]
+
   '@esbuild/android-arm64@0.28.1':
     resolution: {integrity: sha512-34EGEbCIAgosYz6goLcopX6Mo7NyGv9tfwEM2/7Ce2VcVRk568iSvniGWcUXIy7wEDR1wzolcxcriFVrWYcwBg==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [android]
 
+  '@esbuild/android-arm@0.25.4':
+    resolution: {integrity: sha512-QNdQEps7DfFwE3hXiU4BZeOV68HHzYwGd0Nthhd3uCkkEKK7/R6MTgM0P7H7FAs5pU/DIWsviMmEGxEoxIZ+ZQ==}
+    engines: {node: '>=18'}
+    cpu: [arm]
+    os: [android]
+
   '@esbuild/android-arm@0.28.1':
     resolution: {integrity: sha512-0k2F129Xdio1TdJfzJ8sy1Q47vUD2NnwdhiAf7drUN1EBTfPf4hsFCtmMgu/6m8JSzsBrlmVjudMBQqOfG8usQ==}
     engines: {node: '>=18'}
     cpu: [arm]
     os: [android]
 
+  '@esbuild/android-x64@0.25.4':
+    resolution: {integrity: sha512-TVhdVtQIFuVpIIR282btcGC2oGQoSfZfmBdTip2anCaVYcqWlZXGcdcKIUklfX2wj0JklNYgz39OBqh2cqXvcQ==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [android]
+
   '@esbuild/android-x64@0.28.1':
     resolution: {integrity: sha512-dbwY7ltSMDWsRatcRpCnES4F+im88OCUgGZjy52shC7GqHRE/cYlxNbB4Z4UpJswpcc4Qxd2oE/ufM0p61IKng==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [android]
 
+  '@esbuild/darwin-arm64@0.25.4':
+    resolution: {integrity: sha512-Y1giCfM4nlHDWEfSckMzeWNdQS31BQGs9/rouw6Ub91tkK79aIMTH3q9xHvzH8d0wDru5Ci0kWB8b3up/nl16g==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [darwin]
+
   '@esbuild/darwin-arm64@0.28.1':
     resolution: {integrity: sha512-TZbWkQY7kvTAXbXUT7uVACR5cMHsDiSz9z7ZKAX/RTq/WJEk3QyRr0wZpNhBDX+/0CtdqUIJlOiodQcta6tY3Q==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [darwin]
 
+  '@esbuild/darwin-x64@0.25.4':
+    resolution: {integrity: sha512-CJsry8ZGM5VFVeyUYB3cdKpd/H69PYez4eJh1W/t38vzutdjEjtP7hB6eLKBoOdxcAlCtEYHzQ/PJ/oU9I4u0A==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [darwin]
+
   '@esbuild/darwin-x64@0.28.1':
     resolution: {integrity: sha512-zfdzgK9ACBNZLI/CyHTOx81SyNbM6YXn7rxSgX97VjyiPl9W1i4Ka4fgKECEoFCKGpvBj5qArWIGgQjOwkgskQ==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [darwin]
 
+  '@esbuild/freebsd-arm64@0.25.4':
+    resolution: {integrity: sha512-yYq+39NlTRzU2XmoPW4l5Ifpl9fqSk0nAJYM/V/WUGPEFfek1epLHJIkTQM6bBs1swApjO5nWgvr843g6TjxuQ==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [freebsd]
+
   '@esbuild/freebsd-arm64@0.28.1':
     resolution: {integrity: sha512-wG2EA8ENdEI0qhkSZMjfqrdY+ziCYCPMmtZjjIwOmXFjmyzEHn+UUxk5of+SYsjtfs3VpnlC7QLzSI5hY/rOAw==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [freebsd]
 
+  '@esbuild/freebsd-x64@0.25.4':
+    resolution: {integrity: sha512-0FgvOJ6UUMflsHSPLzdfDnnBBVoCDtBTVyn/MrWloUNvq/5SFmh13l3dvgRPkDihRxb77Y17MbqbCAa2strMQQ==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [freebsd]
+
   '@esbuild/freebsd-x64@0.28.1':
     resolution: {integrity: sha512-i7dZ9vQgnvSCzi/rYCXNgtF/U+eKZNJBzu3eTQbRgHnM7tNSizLOkRFAl3qzVc/Op/u5YkHHa4pf/3DOYHthLQ==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [freebsd]
 
+  '@esbuild/linux-arm64@0.25.4':
+    resolution: {integrity: sha512-+89UsQTfXdmjIvZS6nUnOOLoXnkUTB9hR5QAeLrQdzOSWZvNSAXAtcRDHWtqAUtAmv7ZM1WPOOeSxDzzzMogiQ==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [linux]
+
   '@esbuild/linux-arm64@0.28.1':
     resolution: {integrity: sha512-yHs+0uc8+nvEAfAfxrWQKK5peSNzBc4PegcMO0EJ2hT71uA7vB8Ihg2e77R2P7SG5uYjPbHlLLmve4LLLRCf0g==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [linux]
 
+  '@esbuild/linux-arm@0.25.4':
+    resolution: {integrity: sha512-kro4c0P85GMfFYqW4TWOpvmF8rFShbWGnrLqlzp4X1TNWjRY3JMYUfDCtOxPKOIY8B0WC8HN51hGP4I4hz4AaQ==}
+    engines: {node: '>=18'}
+    cpu: [arm]
+    os: [linux]
+
   '@esbuild/linux-arm@0.28.1':
     resolution: {integrity: sha512-qVXBOHQS+d5Y722GwJzJUtOLlX7km3CraOaGormF1pDtPd2C/l1SHRPgjLunLGe51Sh5YYWKMFDyV4SxgMQYTQ==}
     engines: {node: '>=18'}
     cpu: [arm]
     os: [linux]
 
+  '@esbuild/linux-ia32@0.25.4':
+    resolution: {integrity: sha512-yTEjoapy8UP3rv8dB0ip3AfMpRbyhSN3+hY8mo/i4QXFeDxmiYbEKp3ZRjBKcOP862Ua4b1PDfwlvbuwY7hIGQ==}
+    engines: {node: '>=18'}
+    cpu: [ia32]
+    os: [linux]
+
   '@esbuild/linux-ia32@0.28.1':
     resolution: {integrity: sha512-d1z4ZuP0ajrfz/FhGT4vv278rX8KnPPJx8i5+AtK7TYbx9Le9F1hyzurZpkEyjkGa9dUGhQow4C1NmeGvqxN2w==}
     engines: {node: '>=18'}
     cpu: [ia32]
     os: [linux]
 
+  '@esbuild/linux-loong64@0.25.4':
+    resolution: {integrity: sha512-NeqqYkrcGzFwi6CGRGNMOjWGGSYOpqwCjS9fvaUlX5s3zwOtn1qwg1s2iE2svBe4Q/YOG1q6875lcAoQK/F4VA==}
+    engines: {node: '>=18'}
+    cpu: [loong64]
+    os: [linux]
+
   '@esbuild/linux-loong64@0.28.1':
     resolution: {integrity: sha512-M5sRjUVZrkm1OAPR3dlOYzNmN+loZKGVi1VUQGrwuqLcbR6qeAz+famMhjASeH3YVKvZz+zT1jlh/keC3Rj/lg==}
     engines: {node: '>=18'}
     cpu: [loong64]
     os: [linux]
 
+  '@esbuild/linux-mips64el@0.25.4':
+    resolution: {integrity: sha512-IcvTlF9dtLrfL/M8WgNI/qJYBENP3ekgsHbYUIzEzq5XJzzVEV/fXY9WFPfEEXmu3ck2qJP8LG/p3Q8f7Zc2Xg==}
+    engines: {node: '>=18'}
+    cpu: [mips64el]
+    os: [linux]
+
   '@esbuild/linux-mips64el@0.28.1':
     resolution: {integrity: sha512-mRObBZeHh2OxcBFPWE/FjylkRgZdYuiTR3vaTozquCGOH14iP9oN4x4Ge81CoIDYQrXmIxpFumJBu5MtZpnQJQ==}
     engines: {node: '>=18'}
     cpu: [mips64el]
     os: [linux]
 
+  '@esbuild/linux-ppc64@0.25.4':
+    resolution: {integrity: sha512-HOy0aLTJTVtoTeGZh4HSXaO6M95qu4k5lJcH4gxv56iaycfz1S8GO/5Jh6X4Y1YiI0h7cRyLi+HixMR+88swag==}
+    engines: {node: '>=18'}
+    cpu: [ppc64]
+    os: [linux]
+
   '@esbuild/linux-ppc64@0.28.1':
     resolution: {integrity: sha512-slScBsMAb3GFDcdrCgLwZtPYRoH2H/youv10QiZyRjmsP48fznoveWytSgCI/R0ZcUgpc0ZhIUEx6LHts8yrfQ==}
     engines: {node: '>=18'}
     cpu: [ppc64]
     os: [linux]
 
+  '@esbuild/linux-riscv64@0.25.4':
+    resolution: {integrity: sha512-i8JUDAufpz9jOzo4yIShCTcXzS07vEgWzyX3NH2G7LEFVgrLEhjwL3ajFE4fZI3I4ZgiM7JH3GQ7ReObROvSUA==}
+    engines: {node: '>=18'}
+    cpu: [riscv64]
+    os: [linux]
+
   '@esbuild/linux-riscv64@0.28.1':
     resolution: {integrity: sha512-kw0owk1o0GFETUJyW0jc0G4Yzs0BHZn0JDZ8JRT088vjJYX777BAs1fDGxAC+q831qOs2DTC96mNsG2opdfyyQ==}
     engines: {node: '>=18'}
     cpu: [riscv64]
     os: [linux]
 
+  '@esbuild/linux-s390x@0.25.4':
+    resolution: {integrity: sha512-jFnu+6UbLlzIjPQpWCNh5QtrcNfMLjgIavnwPQAfoGx4q17ocOU9MsQ2QVvFxwQoWpZT8DvTLooTvmOQXkO51g==}
+    engines: {node: '>=18'}
+    cpu: [s390x]
+    os: [linux]
+
   '@esbuild/linux-s390x@0.28.1':
     resolution: {integrity: sha512-/lAIjX8aYFRByhh6L5rYtPEDRqa9de/4V/juOXcta5frjvzXO4/sqEtyytse0g3zZFuWu5cDN0MkLz2qRDD2Ag==}
     engines: {node: '>=18'}
     cpu: [s390x]
     os: [linux]
 
+  '@esbuild/linux-x64@0.25.4':
+    resolution: {integrity: sha512-6e0cvXwzOnVWJHq+mskP8DNSrKBr1bULBvnFLpc1KY+d+irZSgZ02TGse5FsafKS5jg2e4pbvK6TPXaF/A6+CA==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [linux]
+
   '@esbuild/linux-x64@0.28.1':
     resolution: {integrity: sha512-u/anNYF2mmVOEDwLtnQ1wOr3EZ9sTNGLWrsYGYwHWzGA3Si84IOkHXlbWTD1NB+9/1lcnweYKO54uhxZydNzfA==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [linux]
 
+  '@esbuild/netbsd-arm64@0.25.4':
+    resolution: {integrity: sha512-vUnkBYxZW4hL/ie91hSqaSNjulOnYXE1VSLusnvHg2u3jewJBz3YzB9+oCw8DABeVqZGg94t9tyZFoHma8gWZQ==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [netbsd]
+
   '@esbuild/netbsd-arm64@0.28.1':
     resolution: {integrity: sha512-oks0DYbLwWMmaakTsCb+zL4E+aHRVLom9IJZOAthMQEPiQmydXHkziYEsGYRx0uNV/IjEKGAV941JzH02pflqw==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [netbsd]
 
+  '@esbuild/netbsd-x64@0.25.4':
+    resolution: {integrity: sha512-XAg8pIQn5CzhOB8odIcAm42QsOfa98SBeKUdo4xa8OvX8LbMZqEtgeWE9P/Wxt7MlG2QqvjGths+nq48TrUiKw==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [netbsd]
+
   '@esbuild/netbsd-x64@0.28.1':
     resolution: {integrity: sha512-aeL6lAnN89Hz43Mlh1G8ARasbuoYvSITDEx0tHh5b7jJnHcssqgjy9Yx430GDpmCa6OyrKoS0aNRjKundRizGg==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [netbsd]
 
+  '@esbuild/openbsd-arm64@0.25.4':
+    resolution: {integrity: sha512-Ct2WcFEANlFDtp1nVAXSNBPDxyU+j7+tId//iHXU2f/lN5AmO4zLyhDcpR5Cz1r08mVxzt3Jpyt4PmXQ1O6+7A==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [openbsd]
+
   '@esbuild/openbsd-arm64@0.28.1':
     resolution: {integrity: sha512-MEFJe5C3R8pwXdZ5Y21oo6m7ePiS0d9pWucn99O/wvyJZChoIQKrQDxKrGeW8F5+T0okTHesAmDeiHDTIq0V/Q==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [openbsd]
 
+  '@esbuild/openbsd-x64@0.25.4':
+    resolution: {integrity: sha512-xAGGhyOQ9Otm1Xu8NT1ifGLnA6M3sJxZ6ixylb+vIUVzvvd6GOALpwQrYrtlPouMqd/vSbgehz6HaVk4+7Afhw==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [openbsd]
+
   '@esbuild/openbsd-x64@0.28.1':
     resolution: {integrity: sha512-i/ZLIOafE0Z8cI/XANJAixoJL/uRAoS2xOA3rb0xN+KK0K177cMAsQYkzHtBrtMXAKuAc7HGgcWiZ/sRC1Nxgw==}
     engines: {node: '>=18'}
@@ -1289,24 +1430,48 @@ packages:
     cpu: [arm64]
     os: [openharmony]
 
+  '@esbuild/sunos-x64@0.25.4':
+    resolution: {integrity: sha512-Mw+tzy4pp6wZEK0+Lwr76pWLjrtjmJyUB23tHKqEDP74R3q95luY/bXqXZeYl4NYlvwOqoRKlInQialgCKy67Q==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [sunos]
+
   '@esbuild/sunos-x64@0.28.1':
     resolution: {integrity: sha512-BEjgtECkL3vY+SaSQ6nzVfiALUeFxpawyp8Jmf5PtYhf1Ug40N1h/hxlhts+f1FvSvarEigdxS3BlSMI2PJLcQ==}
     engines: {node: '>=18'}
     cpu: [x64]
     os: [sunos]
 
+  '@esbuild/win32-arm64@0.25.4':
+    resolution: {integrity: sha512-AVUP428VQTSddguz9dO9ngb+E5aScyg7nOeJDrF1HPYu555gmza3bDGMPhmVXL8svDSoqPCsCPjb265yG/kLKQ==}
+    engines: {node: '>=18'}
+    cpu: [arm64]
+    os: [win32]
+
   '@esbuild/win32-arm64@0.28.1':
     resolution: {integrity: sha512-lCv9eK/H6ZJWbE7bh2nw54CZ9M2nupBxJcTsdk/QQnWkdSjKGuxmmH8/GWrlT1eMmZfn4dGcCjRte397WqfQXA==}
     engines: {node: '>=18'}
     cpu: [arm64]
     os: [win32]
 
+  '@esbuild/win32-ia32@0.25.4':
+    resolution: {integrity: sha512-i1sW+1i+oWvQzSgfRcxxG2k4I9n3O9NRqy8U+uugaT2Dy7kLO9Y7wI72haOahxceMX8hZAzgGou1FhndRldxRg==}
+    engines: {node: '>=18'}
+    cpu: [ia32]
+    os: [win32]
+
   '@esbuild/win32-ia32@0.28.1':
     resolution: {integrity: sha512-zvb/mB2bSCoJOpoCBgYKKpX6YM6mJBlBUVUtVj41DlZJVEB6/0CKlRYxP5wWl1C1ILiCoAU5wZZ4q1P3qeS6Eg==}
     engines: {node: '>=18'}
     cpu: [ia32]
     os: [win32]
 
+  '@esbuild/win32-x64@0.25.4':
+    resolution: {integrity: sha512-nOT2vZNw6hJ+z43oP1SPea/G/6AbN6X+bGNhNuq8NtRHy4wsMhw765IKLNmnjek7GvjWBYQ8Q5VBoYTFg9y1UQ==}
+    engines: {node: '>=18'}
+    cpu: [x64]
+    os: [win32]
+
   '@esbuild/win32-x64@0.28.1':
     resolution: {integrity: sha512-bm4Mowrv+GXMlpWX++EcXw/iLyd1o3+bJkC2DkWXYVvgZCqD/bSj9ctZeAMC3cIxgjRVR2Dufaiu4YPxr5gW1A==}
     engines: {node: '>=18'}
@@ -1730,56 +1895,48 @@ packages:
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [glibc]
 
   '@oxfmt/binding-linux-arm64-musl@0.52.0':
     resolution: {integrity: sha512-wZg6bLjDvh2KibyI3QFUYo8GTXneIFsd0JvehtvJiUmQ8WRPERgxd/VM4ctWb86U5FT1FkqgS8/wZKVB+AZScg==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [musl]
 
   '@oxfmt/binding-linux-ppc64-gnu@0.52.0':
     resolution: {integrity: sha512-IngE8uxhNvxcMrLjZNDo9xNLY7rEK33AKnaMd2B46he1e/mz2CfcW6If/U1wUjdRZddm1QzQaciqZkuMkdh1FA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [ppc64]
     os: [linux]
-    libc: [glibc]
 
   '@oxfmt/binding-linux-riscv64-gnu@0.52.0':
     resolution: {integrity: sha512-H3+DdFMv/efN3Efmhsv18jDrpiWWqKG7wsfAlQBqAt6z/E2Bx+TwEj2Nowe51CPOWB8/mFBC2dAMSgVFLvvowA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [riscv64]
     os: [linux]
-    libc: [glibc]
 
   '@oxfmt/binding-linux-riscv64-musl@0.52.0':
     resolution: {integrity: sha512-zji+1kb7lJKohSDjzC1IsS+K/cKRs1hdVf0ZH0VbdbiakmtLvN9twBoXo/k8VdjFax7kfo+DyPxS7vv52br1aw==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [riscv64]
     os: [linux]
-    libc: [musl]
 
   '@oxfmt/binding-linux-s390x-gnu@0.52.0':
     resolution: {integrity: sha512-hcLBYedpCy7ToUvvBidWk7+11Yhg1oAZ4+6hKPic/mQI6NaqXJSXMps5nFlwUuX2ewhtLZZDPg63TI042qGKBg==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [s390x]
     os: [linux]
-    libc: [glibc]
 
   '@oxfmt/binding-linux-x64-gnu@0.52.0':
     resolution: {integrity: sha512-IDO2loXK2OtTOhSPchU9MW25mWL2QCDGdJbjN8MXKZVS80qXe5gMTwQWu/gMJ3juoBHbkuUZNB2N1LHzNT7DoA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [glibc]
 
   '@oxfmt/binding-linux-x64-musl@0.52.0':
     resolution: {integrity: sha512-mAV2Hjn0SatJ+KoAzKUC3eJhdJ8wv+3m1KyuS0dTsbF0c5weq+QrCt/DRZZM+uj/XiKzCDEUKYsBF30e2qkcyw==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [musl]
 
   '@oxfmt/binding-openharmony-arm64@0.52.0':
     resolution: {integrity: sha512-vd4npaUIwChxp7XzkqmepBWTT9YMcSe/NBApVGPC30/lLyOVaV3dvma1SKo03t8O73BPRAG7EyJzGlN5cJM5hQ==}
@@ -1852,56 +2009,48 @@ packages:
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [glibc]
 
   '@oxlint/binding-linux-arm64-musl@1.71.0':
     resolution: {integrity: sha512-fJZrs5sDZtTaPIOiemRQQmo82Ezy+vOGXemPc4Ok7iVVsYsFa7SlW6Z5XN819VfsqBHRm3NJ3rTdnR8+bJYJdQ==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [musl]
 
   '@oxlint/binding-linux-ppc64-gnu@1.71.0':
     resolution: {integrity: sha512-cwl7VKGERIy9p+G+AvZdfy/06q0aHXaTt/mMRReC751iuNYJgqKjB7NydXSS30nBT9vtr2tunciOtrR4fD6FUA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [ppc64]
     os: [linux]
-    libc: [glibc]
 
   '@oxlint/binding-linux-riscv64-gnu@1.71.0':
     resolution: {integrity: sha512-eZ8ieVXvzGi8jr7+ybQGPK2STw3mldfxZlgA2738iflfB/rzA69sE6m5rDRpQaxC7dpm745Enlh1Tod0QAk9Gg==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [riscv64]
     os: [linux]
-    libc: [glibc]
 
   '@oxlint/binding-linux-riscv64-musl@1.71.0':
     resolution: {integrity: sha512-puMDbQYe6+NXwfMusojoA7CXGn2b3utukmd23PQqc1E3XhVCwyZ+FueSMzDYeNgDV2dUfIVXAAKZBcFDeCL6sA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [riscv64]
     os: [linux]
-    libc: [musl]
 
   '@oxlint/binding-linux-s390x-gnu@1.71.0':
     resolution: {integrity: sha512-4NJLxBs1ujISCt3L/1FcywLs73PWtJuw+piD6feK2V6h6OS6P7xu9/sWt1DTRLibe6QCzmfZzmM/2HPORoV/Lg==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [s390x]
     os: [linux]
-    libc: [glibc]
 
   '@oxlint/binding-linux-x64-gnu@1.71.0':
     resolution: {integrity: sha512-cFDaiR8L3430qp88tfZnvFlt3KotFhR/DlbIL0nHOMMYiG/9Wy4l+6f7t8G8pTa9bd8Lt8+M0y/qjRQ/xcB74g==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [glibc]
 
   '@oxlint/binding-linux-x64-musl@1.71.0':
     resolution: {integrity: sha512-orfixdt76KlpNly9z0PkWBBNfwjKz+JFVLP/7wnVchlKNU9Dpt9InU/ZggeSej6fC7qwHmHNOGlhLnQXcYoGuA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [musl]
 
   '@oxlint/binding-openharmony-arm64@1.71.0':
     resolution: {integrity: sha512-9emQu2lAp6yhPB3XuI+++vR+l/o6JR1X+EpxwcumPdQXBWXEPAsquPGL7l158EqU8SebQMXTUa/S5zN98juyHw==}
@@ -2363,42 +2512,36 @@ packages:
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [glibc]
 
   '@rolldown/binding-linux-arm64-musl@1.1.3':
     resolution: {integrity: sha512-BO9+oPL8K9poZJBfYPsXNtYjPE5uM3qeehT3aFcW4LITOl+iSqhp0abzjR2nWBUNjIZeKXjAEWBZ64WjNoHd6w==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [arm64]
     os: [linux]
-    libc: [musl]
 
   '@rolldown/binding-linux-ppc64-gnu@1.1.3':
     resolution: {integrity: sha512-f3VpLB1vQ0Eo6ecr/6cekLnvYMFF4YBFoVGkfkvPLq1bAkbAwHYQPZKoAmG6OJyTcxxoC+AvezGx/S1obNC0Mw==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [ppc64]
     os: [linux]
-    libc: [glibc]
 
   '@rolldown/binding-linux-s390x-gnu@1.1.3':
     resolution: {integrity: sha512-AmurZ26Pqx/RI9N1gzEOCklkKXl927yjfXWUUS0O7Puh8ARM/Ob8qfrD3qnWksScdw6cSrW5PSHE9DyLu7+PtA==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [s390x]
     os: [linux]
-    libc: [glibc]
 
   '@rolldown/binding-linux-x64-gnu@1.1.3':
     resolution: {integrity: sha512-JJpqs8bRGITDOdbkNKnlojzBabbOHrqjSvDr0IVsZObE1lBcPjxItUEY9eWIDbxaJ3cGrXPWGfGkIxFijg/URg==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [glibc]
 
   '@rolldown/binding-linux-x64-musl@1.1.3':
     resolution: {integrity: sha512-rSJcdjPxzA/by/6/rYs+v+bXU7UjvnbUWz8MJb6kh6+knqB1dCrtHg0uu7C/4haqJvqdkYHQ5IGn+tCH9GLW/g==}
     engines: {node: ^20.19.0 || >=22.12.0}
     cpu: [x64]
     os: [linux]
-    libc: [musl]
 
   '@rolldown/binding-openharmony-arm64@1.1.3':
     resolution: {integrity: sha512-hQ3/PYkDJICgevvyNcVrihVeqq7k1Pp3VZ9lY+dauAYUJKO+auqApvANhvR1An9BhmqYKvW2Mu1F9u4DXSMLxQ==}
@@ -2653,6 +2796,19 @@ packages:
     resolution: {integrity: sha512-A9gOqLdi6cV4ibazAjcQufGj0B1y/vDqYrcuP6d/6x8P27gRS8643Dj9o1dEKtB6O7fwxb2FgBmJS2mX7gpvdw==}
     engines: {node: '>=14.6'}
 
+  '@xterm/addon-unicode11@0.10.0-beta.285':
+    resolution: {integrity: sha512-rfijFu7UcYpaFx5wzxvTpQbIyyq/amf2PuS9pktywcFQr4ITxRgid5EVzKLRG1vchkApNcQplWeYxGEtjiw0Cw==}
+    peerDependencies:
+      '@xterm/xterm': ^6.1.0-beta.285
+
+  '@xterm/addon-webgl@0.20.0-beta.284':
+    resolution: {integrity: sha512-tzkUiEfdpHCY8mXCbuIaP9V67QDfBJvDr9jdxs5jjxNCIQvw+NCoKD97y5sUrQhrIlr7xrDGniPgPYThQ/1FWg==}
+    peerDependencies:
+      '@xterm/xterm': ^6.1.0-beta.285
+
+  '@xterm/xterm@6.1.0-beta.285':
+    resolution: {integrity: sha512-S3K58tepMkbpWRBzOGKd0In6AVvt9QPAnNs8DJ8rPUPODYtsCYWAtINHKYtC2OpXcE5EBKM35dl+Dgv03OoE/w==}
+
   abab@2.0.6:
     resolution: {integrity: sha512-j2afSsaIENvHZN2B8GOpF566vZ5WVk5opAiMTvWgaQT8DkbOqsTfvNAvHoRGU2zzP8cPoqys+xHTRDWW8L+/BA==}
     deprecated: Use your platform's native atob() and btoa() methods instead
@@ -2681,8 +2837,8 @@ packages:
     resolution: {integrity: sha512-HEHNfbars9v4pgpW6SO1KSPkfoS0xVOM/9UzkJltjlsHZmJasxg8aXkuZa7SMf8vKGIBhpUsPluQSqhJFCqebw==}
     engines: {node: '>=0.4.0'}
 
-  acorn@8.16.0:
-    resolution: {integrity: sha512-UVJyE9MttOsBQIDKw1skb9nAwQuR5wuGD3+82K6JgJlm/Y+KI92oNsMNGZCYdDsVtRHSak0pcV5Dno5+4jh9sw==}
+  acorn@8.15.0:
+    resolution: {integrity: sha512-NZyJarBfL7nWwIq+FDL6Zp/yHEhePMNnnJ0y3qfieCrmNvYct8uvtiV41UvlSe6apAfk0fY1FbWx+NwfmpvtTg==}
     engines: {node: '>=0.4.0'}
     hasBin: true
 
@@ -3378,6 +3534,11 @@ packages:
     resolution: {integrity: sha512-w+5mJ3GuFL+NjVtJlvydShqE1eN3h3PbI7/5LAsYJP/2qtuMXjfL2LpHSRqo4b4eSF5K/DH1JXKUAHSB2UW50g==}
     engines: {node: '>= 0.4'}
 
+  esbuild@0.25.4:
+    resolution: {integrity: sha512-8pgjLUcUjcgDg+2Q4NYXnPbo/vncAY4UmyaCm0jZevERqCHZIaWwdJHkf8XQtu4AxSKCdvrUbT0XUr1IdZzI8Q==}
+    engines: {node: '>=18'}
+    hasBin: true
+
   esbuild@0.28.1:
     resolution: {integrity: sha512-HrJrvZv5ayxBzPfwphOoNzkzOIIlifzk0KJrGK2c8R4+LKpMtpYLQeUdjnwjWv/LZlkH2laZk+4w78pi99D4Vw==}
     engines: {node: '>=18'}
@@ -4660,28 +4821,24 @@ packages:
     engines: {node: '>= 12.0.0'}
     cpu: [arm64]
     os: [linux]
-    libc: [glibc]
 
   lightningcss-linux-arm64-musl@1.32.0:
     resolution: {integrity: sha512-UpQkoenr4UJEzgVIYpI80lDFvRmPVg6oqboNHfoH4CQIfNA+HOrZ7Mo7KZP02dC6LjghPQJeBsvXhJod/wnIBg==}
     engines: {node: '>= 12.0.0'}
     cpu: [arm64]
     os: [linux]
-    libc: [musl]
 
   lightningcss-linux-x64-gnu@1.32.0:
     resolution: {integrity: sha512-V7Qr52IhZmdKPVr+Vtw8o+WLsQJYCTd8loIfpDaMRWGUZfBOYEJeyJIkqGIDMZPwPx24pUMfwSxxI8phr/MbOA==}
     engines: {node: '>= 12.0.0'}
     cpu: [x64]
     os: [linux]
-    libc: [glibc]
 
   lightningcss-linux-x64-musl@1.32.0:
     resolution: {integrity: sha512-bYcLp+Vb0awsiXg/80uCRezCYHNg1/l3mt0gzHnWV9XP1W5sKa5/TCdGWaR/zBM2PeF/HbsQv/j2URNOiVuxWg==}
     engines: {node: '>= 12.0.0'}
     cpu: [x64]
     os: [linux]
-    libc: [musl]
 
   lightningcss-win32-arm64-msvc@1.32.0:
     resolution: {integrity: sha512-8SbC8BR40pS6baCM8sbtYDSwEVQd4JlFTOlaD3gWGHfThTcABnNDBda6eTZeqbofalIJhFx0qKzgHJmcPTnGdw==}
@@ -7675,81 +7832,156 @@ snapshots:
       tslib: 2.8.1
     optional: true
 
+  '@esbuild/aix-ppc64@0.25.4':
+    optional: true
+
   '@esbuild/aix-ppc64@0.28.1':
     optional: true
 
+  '@esbuild/android-arm64@0.25.4':
+    optional: true
+
   '@esbuild/android-arm64@0.28.1':
     optional: true
 
+  '@esbuild/android-arm@0.25.4':
+    optional: true
+
   '@esbuild/android-arm@0.28.1':
     optional: true
 
+  '@esbuild/android-x64@0.25.4':
+    optional: true
+
   '@esbuild/android-x64@0.28.1':
     optional: true
 
+  '@esbuild/darwin-arm64@0.25.4':
+    optional: true
+
   '@esbuild/darwin-arm64@0.28.1':
     optional: true
 
+  '@esbuild/darwin-x64@0.25.4':
+    optional: true
+
   '@esbuild/darwin-x64@0.28.1':
     optional: true
 
+  '@esbuild/freebsd-arm64@0.25.4':
+    optional: true
+
   '@esbuild/freebsd-arm64@0.28.1':
     optional: true
 
+  '@esbuild/freebsd-x64@0.25.4':
+    optional: true
+
   '@esbuild/freebsd-x64@0.28.1':
     optional: true
 
+  '@esbuild/linux-arm64@0.25.4':
+    optional: true
+
   '@esbuild/linux-arm64@0.28.1':
     optional: true
 
+  '@esbuild/linux-arm@0.25.4':
+    optional: true
+
   '@esbuild/linux-arm@0.28.1':
     optional: true
 
+  '@esbuild/linux-ia32@0.25.4':
+    optional: true
+
   '@esbuild/linux-ia32@0.28.1':
     optional: true
 
+  '@esbuild/linux-loong64@0.25.4':
+    optional: true
+
   '@esbuild/linux-loong64@0.28.1':
     optional: true
 
+  '@esbuild/linux-mips64el@0.25.4':
+    optional: true
+
   '@esbuild/linux-mips64el@0.28.1':
     optional: true
 
+  '@esbuild/linux-ppc64@0.25.4':
+    optional: true
+
   '@esbuild/linux-ppc64@0.28.1':
     optional: true
 
+  '@esbuild/linux-riscv64@0.25.4':
+    optional: true
+
   '@esbuild/linux-riscv64@0.28.1':
     optional: true
 
+  '@esbuild/linux-s390x@0.25.4':
+    optional: true
+
   '@esbuild/linux-s390x@0.28.1':
     optional: true
 
+  '@esbuild/linux-x64@0.25.4':
+    optional: true
+
   '@esbuild/linux-x64@0.28.1':
     optional: true
 
+  '@esbuild/netbsd-arm64@0.25.4':
+    optional: true
+
   '@esbuild/netbsd-arm64@0.28.1':
     optional: true
 
+  '@esbuild/netbsd-x64@0.25.4':
+    optional: true
+
   '@esbuild/netbsd-x64@0.28.1':
     optional: true
 
+  '@esbuild/openbsd-arm64@0.25.4':
+    optional: true
+
   '@esbuild/openbsd-arm64@0.28.1':
     optional: true
 
+  '@esbuild/openbsd-x64@0.25.4':
+    optional: true
+
   '@esbuild/openbsd-x64@0.28.1':
     optional: true
 
   '@esbuild/openharmony-arm64@0.28.1':
     optional: true
 
+  '@esbuild/sunos-x64@0.25.4':
+    optional: true
+
   '@esbuild/sunos-x64@0.28.1':
     optional: true
 
+  '@esbuild/win32-arm64@0.25.4':
+    optional: true
+
   '@esbuild/win32-arm64@0.28.1':
     optional: true
 
+  '@esbuild/win32-ia32@0.25.4':
+    optional: true
+
   '@esbuild/win32-ia32@0.28.1':
     optional: true
 
+  '@esbuild/win32-x64@0.25.4':
+    optional: true
+
   '@esbuild/win32-x64@0.28.1':
     optional: true
 
@@ -9327,13 +9559,13 @@ snapshots:
       chai: 6.2.2
       tinyrainbow: 3.1.0
 
-  '@vitest/mocker@4.1.9(vite@8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))':
+  '@vitest/mocker@4.1.9(vite@8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))':
     dependencies:
       '@vitest/spy': 4.1.9
       estree-walker: 3.0.3
       magic-string: 0.30.21
     optionalDependencies:
-      vite: 8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
+      vite: 8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
 
   '@vitest/pretty-format@4.1.9':
     dependencies:
@@ -9363,6 +9595,16 @@ snapshots:
 
   '@xmldom/xmldom@0.9.10': {}
 
+  '@xterm/addon-unicode11@0.10.0-beta.285(@xterm/xterm@6.1.0-beta.285)':
+    dependencies:
+      '@xterm/xterm': 6.1.0-beta.285
+
+  '@xterm/addon-webgl@0.20.0-beta.284(@xterm/xterm@6.1.0-beta.285)':
+    dependencies:
+      '@xterm/xterm': 6.1.0-beta.285
+
+  '@xterm/xterm@6.1.0-beta.285': {}
+
   abab@2.0.6: {}
 
   abort-controller@3.0.0:
@@ -9381,7 +9623,7 @@ snapshots:
 
   acorn-globals@7.0.1:
     dependencies:
-      acorn: 8.16.0
+      acorn: 8.15.0
       acorn-walk: 8.3.5
 
   acorn-jsx@5.3.2(acorn@8.17.0):
@@ -9390,9 +9632,9 @@ snapshots:
 
   acorn-walk@8.3.5:
     dependencies:
-      acorn: 8.16.0
+      acorn: 8.15.0
 
-  acorn@8.16.0: {}
+  acorn@8.15.0: {}
 
   acorn@8.17.0: {}
 
@@ -10235,6 +10477,34 @@ snapshots:
       is-date-object: 1.1.0
       is-symbol: 1.1.1
 
+  esbuild@0.25.4:
+    optionalDependencies:
+      '@esbuild/aix-ppc64': 0.25.4
+      '@esbuild/android-arm': 0.25.4
+      '@esbuild/android-arm64': 0.25.4
+      '@esbuild/android-x64': 0.25.4
+      '@esbuild/darwin-arm64': 0.25.4
+      '@esbuild/darwin-x64': 0.25.4
+      '@esbuild/freebsd-arm64': 0.25.4
+      '@esbuild/freebsd-x64': 0.25.4
+      '@esbuild/linux-arm': 0.25.4
+      '@esbuild/linux-arm64': 0.25.4
+      '@esbuild/linux-ia32': 0.25.4
+      '@esbuild/linux-loong64': 0.25.4
+      '@esbuild/linux-mips64el': 0.25.4
+      '@esbuild/linux-ppc64': 0.25.4
+      '@esbuild/linux-riscv64': 0.25.4
+      '@esbuild/linux-s390x': 0.25.4
+      '@esbuild/linux-x64': 0.25.4
+      '@esbuild/netbsd-arm64': 0.25.4
+      '@esbuild/netbsd-x64': 0.25.4
+      '@esbuild/openbsd-arm64': 0.25.4
+      '@esbuild/openbsd-x64': 0.25.4
+      '@esbuild/sunos-x64': 0.25.4
+      '@esbuild/win32-arm64': 0.25.4
+      '@esbuild/win32-ia32': 0.25.4
+      '@esbuild/win32-x64': 0.25.4
+
   esbuild@0.28.1:
     optionalDependencies:
       '@esbuild/aix-ppc64': 0.28.1
@@ -10688,7 +10958,7 @@ snapshots:
       expo: 55.0.23(@babel/core@7.29.7)(@expo/dom-webview@55.0.5)(@expo/metro-runtime@55.0.10)(expo-router@55.0.14)(react-dom@19.2.6(react@19.2.6))(react-native-webview@13.16.1(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react@19.2.6))(react-native-worklets@0.8.3(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react@19.2.6))(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react@19.2.6)(typescript@5.9.3)
       expo-json-utils: 55.0.2
 
-  expo-module-scripts@55.0.2(@babel/core@7.29.7)(@babel/runtime@7.29.2)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.28.1)(eslint@9.39.4)(expo@55.0.23)(jest@29.7.0(@types/node@25.9.3))(prettier@2.8.8)(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react-refresh@0.14.2)(react-test-renderer@19.2.6(react@19.2.6))(react@19.2.6):
+  expo-module-scripts@55.0.2(@babel/core@7.29.7)(@babel/runtime@7.29.2)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.25.4)(eslint@9.39.4)(expo@55.0.23)(jest@29.7.0(@types/node@25.9.3))(prettier@2.8.8)(react-native@0.83.9(@babel/core@7.29.7)(@react-native/metro-config@0.85.2(@babel/core@7.29.7))(@types/react@19.2.14)(react@19.2.6))(react-refresh@0.14.2)(react-test-renderer@19.2.6(react@19.2.6))(react@19.2.6):
     dependencies:
       '@babel/cli': 7.28.6(@babel/core@7.29.7)
       '@babel/plugin-transform-export-namespace-from': 7.27.1(@babel/core@7.29.7)
@@ -10708,7 +10978,7 @@ snapshots:
       jest-snapshot-prettier: prettier@2.8.8
       jest-watch-typeahead: 2.2.1(jest@29.7.0(@types/node@25.9.3))
       resolve-workspace-root: 2.0.1
-      ts-jest: 29.0.5(@babel/core@7.29.7)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.28.1)(jest@29.7.0(@types/node@25.9.3))(typescript@5.9.3)
+      ts-jest: 29.0.5(@babel/core@7.29.7)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.25.4)(jest@29.7.0(@types/node@25.9.3))(typescript@5.9.3)
       typescript: 5.9.3
     transitivePeerDependencies:
       - '@babel/core'
@@ -11832,7 +12102,7 @@ snapshots:
   jsdom@20.0.3:
     dependencies:
       abab: 2.0.6
-      acorn: 8.16.0
+      acorn: 8.15.0
       acorn-globals: 7.0.1
       cssom: 0.5.0
       cssstyle: 2.3.0
@@ -13463,7 +13733,7 @@ snapshots:
   terser@5.46.2:
     dependencies:
       '@jridgewell/source-map': 0.3.11
-      acorn: 8.16.0
+      acorn: 8.15.0
       commander: 2.20.3
       source-map-support: 0.5.21
 
@@ -13527,7 +13797,7 @@ snapshots:
       picomatch: 4.0.4
       typescript: 5.9.3
 
-  ts-jest@29.0.5(@babel/core@7.29.7)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.28.1)(jest@29.7.0(@types/node@25.9.3))(typescript@5.9.3):
+  ts-jest@29.0.5(@babel/core@7.29.7)(@jest/types@29.6.3)(babel-jest@29.7.0(@babel/core@7.29.7))(esbuild@0.25.4)(jest@29.7.0(@types/node@25.9.3))(typescript@5.9.3):
     dependencies:
       bs-logger: 0.2.6
       fast-json-stable-stringify: 2.1.0
@@ -13543,7 +13813,7 @@ snapshots:
       '@babel/core': 7.29.7
       '@jest/types': 29.6.3
       babel-jest: 29.7.0(@babel/core@7.29.7)
-      esbuild: 0.28.1
+      esbuild: 0.25.4
 
   tsconfig-paths@3.15.0:
     dependencies:
@@ -13700,7 +13970,7 @@ snapshots:
       - '@types/react'
       - '@types/react-dom'
 
-  vite@8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0):
+  vite@8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0):
     dependencies:
       lightningcss: 1.32.0
       picomatch: 4.0.4
@@ -13709,16 +13979,16 @@ snapshots:
       tinyglobby: 0.2.17
     optionalDependencies:
       '@types/node': 25.9.3
-      esbuild: 0.28.1
+      esbuild: 0.25.4
       fsevents: 2.3.3
       terser: 5.48.0
       tsx: 4.22.4
       yaml: 2.9.0
 
-  vitest@4.1.9(@types/node@25.9.3)(jsdom@20.0.3)(vite@8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)):
+  vitest@4.1.9(@types/node@25.9.3)(jsdom@20.0.3)(vite@8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)):
     dependencies:
       '@vitest/expect': 4.1.9
-      '@vitest/mocker': 4.1.9(vite@8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))
+      '@vitest/mocker': 4.1.9(vite@8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0))
       '@vitest/pretty-format': 4.1.9
       '@vitest/runner': 4.1.9
       '@vitest/snapshot': 4.1.9
@@ -13735,7 +14005,7 @@ snapshots:
       tinyexec: 1.1.2
       tinyglobby: 0.2.17
       tinyrainbow: 3.1.0
-      vite: 8.1.0(@types/node@25.9.3)(esbuild@0.28.1)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
+      vite: 8.1.0(@types/node@25.9.3)(esbuild@0.25.4)(terser@5.48.0)(tsx@4.22.4)(yaml@2.9.0)
       why-is-node-running: 2.3.0
     optionalDependencies:
       '@types/node': 25.9.3

mobile/scripts/build-terminal-webview-engine.mjs

@@ -0,0 +1,108 @@
+import { readFile, writeFile } from 'node:fs/promises'
+import path from 'node:path'
+import { createRequire } from 'node:module'
+import * as esbuild from 'esbuild'
+
+const require = createRequire(import.meta.url)
+const scriptDir = import.meta.dirname
+const mobileRoot = path.resolve(scriptDir, '..')
+const outputPath = path.join(mobileRoot, 'src', 'terminal', 'terminal-webview-engine.generated.ts')
+const target = 'chrome74'
+
+const packages = ['@xterm/xterm', '@xterm/addon-unicode11', '@xterm/addon-webgl']
+
+async function readPackageVersion(packageName) {
+  // Why: a package.json module specifier must use '/' — path.join emits '\' on
+  // Windows, yielding an unresolvable bare specifier that fails postinstall there.
+  const packageJsonPath = require.resolve(`${packageName}/package.json`)
+  const packageJson = JSON.parse(await readFile(packageJsonPath, 'utf8'))
+  return `${packageName}@${packageJson.version}`
+}
+
+function htmlText(value, closingTag) {
+  return value.replace(new RegExp(`</${closingTag}`, 'gi'), `<\\/${closingTag}`)
+}
+
+async function buildEngineJs() {
+  const result = await esbuild.build({
+    stdin: {
+      contents: `
+        import { Terminal } from '@xterm/xterm'
+        import { Unicode11Addon } from '@xterm/addon-unicode11'
+        import { WebglAddon } from '@xterm/addon-webgl'
+
+        // Why: xterm reaches for these runtime APIs on the terminal-bringup path,
+        // and esbuild lowers syntax but not runtime APIs. Guarded shims let the
+        // chrome74-syntax bundle actually run on old WebViews (the #7030 goal)
+        // instead of throwing at construction and only surfacing the error overlay.
+
+        // WeakRef (Chrome 84+): lazily constructed in window-tracking paths.
+        // Strong retention is fine for a single-document terminal WebView.
+        if (typeof window.WeakRef === 'undefined') {
+          window.WeakRef = function WeakRefShim(target) { this.__target = target }
+          window.WeakRef.prototype.deref = function () { return this.__target }
+        }
+
+        // structuredClone (Chrome 98+): xterm clones its plain-data DEC-mode default
+        // objects at Terminal construction; a JSON round-trip clones those correctly
+        // (any undefined-valued keys drop, but every reader treats absent == undefined).
+        if (typeof window.structuredClone === 'undefined') {
+          window.structuredClone = function (value) { return JSON.parse(JSON.stringify(value)) }
+        }
+
+        // Element.prototype.replaceChildren (Chrome 86+): used on the row/selection
+        // render path; polyfill via remove-all + append (Chrome 54+, under the floor).
+        if (typeof Element !== 'undefined' && !Element.prototype.replaceChildren) {
+          Element.prototype.replaceChildren = function () {
+            while (this.firstChild) this.removeChild(this.firstChild)
+            this.append.apply(this, arguments)
+          }
+        }
+
+        window.Terminal = Terminal
+        window.Unicode11Addon = { Unicode11Addon }
+        window.WebglAddon = { WebglAddon }
+      `,
+      resolveDir: mobileRoot,
+      sourcefile: 'terminal-webview-engine-entry.js'
+    },
+    bundle: true,
+    format: 'iife',
+    minify: true,
+    platform: 'browser',
+    target,
+    legalComments: 'none',
+    write: false,
+    logLevel: 'silent'
+  })
+
+  return result.outputFiles[0].text
+}
+
+async function main() {
+  const [engineJs, rawEngineCss, ...versions] = await Promise.all([
+    buildEngineJs(),
+    readFile(require.resolve('@xterm/xterm/css/xterm.css'), 'utf8'),
+    ...packages.map(readPackageVersion)
+  ])
+  // Why: the no-external-URL regression gate bans http(s):// anywhere in the
+  // terminal document. These xmlns URIs live inside data: URLs (never fetched);
+  // percent-encoding the scheme colon satisfies the gate and URI-decodes back
+  // before the SVG is parsed.
+  const engineCss = rawEngineCss
+    .replace(/\/\*[\s\S]*?\*\//g, '')
+    .replace(/http:\/\/www\.w3\.org\/2000\/svg/g, 'http%3A//www.w3.org/2000/svg')
+
+  const source = [
+    '// Generated by scripts/build-terminal-webview-engine.mjs.',
+    `// Packages: ${versions.join(', ')}.`,
+    `// Target: ${target}. Do not edit by hand; regenerate via pnpm postinstall.`,
+    `export const XTERM_ENGINE_JS = ${JSON.stringify(htmlText(engineJs, 'script'))}`,
+    `export const XTERM_ENGINE_CSS = ${JSON.stringify(htmlText(engineCss, 'style'))}`,
+    ''
+  ].join('\n')
+
+  await writeFile(outputPath, source)
+}
+
+await main()

mobile/scripts/mock-server-file-preview-data.ts

@@ -21,6 +21,31 @@ const MOCK_FILE_LIST = [
   { relativePath: 'dist/archive.zip', basename: 'archive.zip', kind: 'binary' }
 ]
 
+function readMockDirectory(relativePath: string): Array<{ name: string; isDirectory: boolean }> {
+  const prefix = relativePath ? `${relativePath}/` : ''
+  const children = new Map<string, boolean>()
+  for (const file of MOCK_FILE_LIST) {
+    if (!file.relativePath.startsWith(prefix)) {
+      continue
+    }
+    const rest = file.relativePath.slice(prefix.length)
+    if (!rest) {
+      continue
+    }
+    const [name, ...descendants] = rest.split('/')
+    if (!name) {
+      continue
+    }
+    children.set(name, children.get(name) === true || descendants.length > 0)
+  }
+  return Array.from(children, ([name, isDirectory]) => ({ name, isDirectory })).sort((a, b) => {
+    if (a.isDirectory !== b.isDirectory) {
+      return a.isDirectory ? -1 : 1
+    }
+    return a.name.localeCompare(b.name)
+  })
+}
+
 export function handleMockFilePreviewRequest(
   request: RpcRequest,
   respond: Respond,
@@ -40,6 +65,10 @@ export function handleMockFilePreviewRequest(
       )
       return true
 
+    case 'files.readDir':
+      respond(success(request.id, readMockDirectory(String(request.params?.relativePath ?? ''))))
+      return true
+
     case 'files.read': {
       const relativePath = String(request.params?.relativePath ?? '')
       const content = MOCK_FILE_CONTENT[relativePath]

mobile/scripts/prepare-android-release.mjs

@@ -22,31 +22,12 @@ function fail(message) {
   process.exit(1)
 }
 
-function parsePositiveInteger(value, name) {
-  if (!/^\d+$/.test(value)) {
-    fail(`${name} must be a positive integer`)
-  }
-
-  const parsed = Number(value)
-  if (!Number.isSafeInteger(parsed) || parsed <= 0) {
-    fail(`${name} must be a positive integer`)
-  }
-
-  return parsed
-}
-
 function validateSemver(version, name) {
   if (!semverPattern.test(version)) {
     fail(`${name} must use x.y.z format`)
   }
 }
 
-function bumpPatchVersion(version) {
-  validateSemver(version, 'Current mobile version')
-  const [major, minor, patch] = version.split('.')
-  return `${major}.${minor}.${Number(patch) + 1}`
-}
-
 function writeOutput(name, value) {
   const outputPath = process.env.GITHUB_OUTPUT
   if (!outputPath) {
@@ -74,10 +55,6 @@ const tagVersion = githubRef.startsWith(androidTagRefPrefix)
 const requestedVersion = input('MOBILE_ANDROID_RELEASE_VERSION')
 const bumpPatch = truthy(input('MOBILE_ANDROID_BUMP_PATCH_VERSION'))
 
-if (requestedVersion && bumpPatch) {
-  fail('Use either MOBILE_ANDROID_RELEASE_VERSION or MOBILE_ANDROID_BUMP_PATCH_VERSION, not both')
-}
-
 if (tagVersion) {
   validateSemver(tagVersion, 'Android release tag version')
 }
@@ -86,43 +63,36 @@ if (requestedVersion) {
   validateSemver(requestedVersion, 'MOBILE_ANDROID_RELEASE_VERSION')
 }
 
-if (tagVersion && requestedVersion && requestedVersion !== tagVersion) {
-  fail('MOBILE_ANDROID_RELEASE_VERSION must match the mobile-android-v tag')
+if (bumpPatch) {
+  fail('MOBILE_ANDROID_BUMP_PATCH_VERSION is no longer supported; commit mobile/app.json first')
 }
 
-if (tagVersion && bumpPatch) {
-  fail('MOBILE_ANDROID_BUMP_PATCH_VERSION is only supported for manual branch runs')
+if (tagVersion && tagVersion !== currentVersion) {
+  fail('Android release tag version must match the committed mobile app version')
 }
 
-const version =
-  requestedVersion || tagVersion || (bumpPatch ? bumpPatchVersion(currentVersion) : currentVersion)
+if (requestedVersion && requestedVersion !== currentVersion) {
+  fail('MOBILE_ANDROID_RELEASE_VERSION must match the committed mobile app version')
+}
 
 const requestedVersionCode = input('MOBILE_ANDROID_VERSION_CODE')
 const bumpVersionCode = truthy(input('MOBILE_ANDROID_BUMP_VERSION_CODE'))
 
-if (requestedVersionCode && bumpVersionCode) {
-  fail('Use either MOBILE_ANDROID_VERSION_CODE or MOBILE_ANDROID_BUMP_VERSION_CODE, not both')
+if (requestedVersionCode || bumpVersionCode) {
+  // Why: Android rejects lower versionCode installs; release-only bumps leave
+  // committed dev builds behind shipped APKs and break local testing.
+  fail('Android versionCode changes must be committed in mobile/app.json before release')
 }
 
-const versionCode = requestedVersionCode
-  ? parsePositiveInteger(requestedVersionCode, 'MOBILE_ANDROID_VERSION_CODE')
-  : bumpVersionCode || version !== currentVersion
-    ? currentVersionCode + 1
-    : currentVersionCode
-
-expo.version = version
-android.versionCode = versionCode
-fs.writeFileSync(appConfigPath, `${JSON.stringify(config, null, 2)}\n`)
-
-const tag = `mobile-android-v${version}`
+const tag = `mobile-android-v${currentVersion}`
 const publishRelease =
   githubRef.startsWith(androidTagRefPrefix) || truthy(input('MOBILE_ANDROID_PUBLISH_RELEASE'))
 
-writeOutput('version', version)
-writeOutput('android_version_code', String(versionCode))
+writeOutput('version', currentVersion)
+writeOutput('android_version_code', String(currentVersionCode))
 writeOutput('tag', tag)
 writeOutput('publish_release', publishRelease ? 'true' : 'false')
 
-console.log(`Prepared Orca Mobile Android ${version} (${versionCode})`)
+console.log(`Prepared Orca Mobile Android ${currentVersion} (${currentVersionCode})`)
 console.log(`Release tag: ${tag}`)
 console.log(`Publish GitHub Release: ${publishRelease ? 'yes' : 'no'}`)

mobile/src/components/MobileDiffReviewLine.tsx

@@ -92,9 +92,10 @@ export function MobileDiffReviewLine({
   )
 }
 
+// Row height comes from the 18px code lineHeight alone (no vertical padding or
+// minHeight) so mobile diff density matches the desktop diff editor (STA-1239).
 const styles = StyleSheet.create({
   row: {
-    minHeight: 32,
     flexDirection: 'row',
     alignItems: 'stretch',
     borderBottomWidth: StyleSheet.hairlineWidth,
@@ -112,25 +113,24 @@ const styles = StyleSheet.create({
   },
   prefix: {
     width: 18,
-    paddingTop: spacing.sm,
     textAlign: 'center',
     color: colors.textMuted,
     fontFamily: typography.monoFamily,
-    fontSize: typography.metaSize
+    fontSize: typography.metaSize,
+    lineHeight: 18
   },
   lineNumber: {
     width: 44,
-    paddingTop: spacing.sm,
     paddingRight: spacing.xs,
     textAlign: 'right',
     color: colors.textMuted,
     fontFamily: typography.monoFamily,
-    fontSize: typography.metaSize
+    fontSize: typography.metaSize,
+    lineHeight: 18
   },
   code: {
     flex: 1,
     minWidth: 0,
-    paddingVertical: spacing.sm,
     paddingHorizontal: spacing.sm
   },
   codePressed: {

mobile/src/components/NewWorktreeModal.tsx

@@ -7,7 +7,8 @@ import {
   Switch,
   StyleSheet,
   Platform,
-  ActivityIndicator
+  ActivityIndicator,
+  Keyboard
 } from 'react-native'
 import { ChevronDown, ChevronUp, Check } from 'lucide-react-native'
 import type { RpcClient } from '../transport/rpc-client'
@@ -174,6 +175,7 @@ function NewWorktreeModalContent({
   const [repos, setRepos] = useState<Repo[]>(initialRepos ?? [])
   const [selectedRepo, setSelectedRepo] = useState<Repo | null>(null)
   const [showRepoPicker, setShowRepoPicker] = useState(false)
+  const [nameAutoFocusEnabled, setNameAutoFocusEnabled] = useState(true)
   const [selectedAgentState, setSelectedAgent] = useState<AgentOption>(AGENT_OPTIONS[0]!)
   const [runtimeSettings, setRuntimeSettings] = useState<RuntimeSettings | null>(null)
   const [detectedAgentIdsState, setDetectedAgentIdsState] = useState<DetectedAgentIdsState | null>(
@@ -667,6 +669,13 @@ function NewWorktreeModalContent({
     [repos]
   )
 
+  function prepareSelectionPickerOpen(): void {
+    // Why: picker taps can beat the delayed name-field focus; suppressing it
+    // prevents the keyboard from reopening under the picker drawer.
+    setNameAutoFocusEnabled(false)
+    Keyboard.dismiss()
+  }
+
   return (
     <>
       <BottomDrawer visible={visible} onClose={onClose}>
@@ -689,7 +698,13 @@ function NewWorktreeModalContent({
           <>
             <View style={styles.field}>
               <Text style={styles.label}>Repository</Text>
-              <Pressable style={styles.fieldButton} onPress={() => setShowRepoPicker(true)}>
+              <Pressable
+                style={styles.fieldButton}
+                onPress={() => {
+                  prepareSelectionPickerOpen()
+                  setShowRepoPicker(true)
+                }}
+              >
                 {selectedRepo ? (
                   <View
                     style={[styles.repoDot, { backgroundColor: repoBadgeColor(selectedRepo) }]}
@@ -760,7 +775,7 @@ function NewWorktreeModalContent({
                   setError('')
                 }}
                 placeholderTextColor={colors.textMuted}
-                shouldAutoFocus={visible && !loading && repos.length > 0}
+                shouldAutoFocus={nameAutoFocusEnabled && visible && !loading && repos.length > 0}
                 returnKeyType="done"
                 onSubmitEditing={() => {
                   if (canCreate) {
@@ -775,7 +790,10 @@ function NewWorktreeModalContent({
               <Pressable
                 style={[styles.fieldButton, sshGate.requiresConnection && styles.disabled]}
                 disabled={sshGate.requiresConnection}
-                onPress={() => setShowAgentPicker(true)}
+                onPress={() => {
+                  prepareSelectionPickerOpen()
+                  setShowAgentPicker(true)
+                }}
               >
                 <MobileAgentIcon agentId={selectedAgent.id} size={16} />
                 <Text style={styles.fieldButtonText} numberOfLines={1}>

mobile/src/components/pr-sidebar/MobilePrViewPanel.tsx

@@ -2,13 +2,14 @@ import { useEffect } from 'react'
 import { Pressable, StyleSheet, Text, View } from 'react-native'
 import { SafeAreaView, useSafeAreaInsets } from 'react-native-safe-area-context'
 import { useRouter } from 'expo-router'
-import { ChevronLeft, X } from 'lucide-react-native'
+import { ChevronLeft, ExternalLink, X } from 'lucide-react-native'
 import { colors, radii, spacing, typography } from '../../theme/mobile-theme'
 import type { ConnectionState } from '../../transport/types'
 import type { RpcClient } from '../../transport/rpc-client'
 import type { MobileGitStatusResult } from '../../source-control/mobile-git-status'
 import { useMobilePrSidebarController } from '../../session/use-mobile-pr-sidebar-controller'
 import { MobilePRSidebar } from '../MobilePRSidebar'
+import { openMobilePrUrl } from '../MobilePrComposeSheet'
 
 type Props = {
   client: RpcClient | null
@@ -73,6 +74,27 @@ export function MobilePrViewPanel({
             message: 'Current branch unavailable.'
           } as const)
         : controller.prSidebarState
+  // Why: open-on-host lives in the chrome so the PR URL is always flush-right of
+  // the screen title, even when the body is scrolled past the PR header.
+  const prUrl =
+    sidebarState.kind === 'ready' && sidebarState.data.pr.url ? sidebarState.data.pr.url : null
+  const prNumber = sidebarState.kind === 'ready' ? sidebarState.data.pr.number : null
+  const openPr = prUrl ? () => openMobilePrUrl(prUrl) : undefined
+  const openPrControl = openPr ? (
+    <Pressable
+      style={({ pressed }) => [styles.iconButton, pressed && styles.iconButtonPressed]}
+      onPress={openPr}
+      hitSlop={8}
+      accessibilityRole="link"
+      accessibilityLabel={
+        prNumber != null
+          ? `Open pull request #${prNumber} on the web`
+          : 'Open pull request on the web'
+      }
+    >
+      <ExternalLink size={18} color={colors.textSecondary} strokeWidth={2.2} />
+    </Pressable>
+  ) : null
   const sidebar = (
     <MobilePRSidebar
       state={sidebarState}
@@ -104,6 +126,7 @@ export function MobilePrViewPanel({
             <Text style={styles.title} numberOfLines={1}>
               Pull Request
             </Text>
+            {openPrControl}
           </View>
         </View>
         {sidebar}
@@ -126,6 +149,7 @@ export function MobilePrViewPanel({
           <Text style={styles.title} numberOfLines={1}>
             Pull Request
           </Text>
+          {openPrControl}
         </View>
       </View>
       {sidebar}

mobile/src/components/pr-sidebar/PRSidebarHeader.tsx

@@ -29,8 +29,8 @@ export function PRSidebarHeader({ pr, details, titleAction }: Props) {
   const baseRef = item?.baseRefName ?? null
   const headRef = item?.branchName ?? null
   const editable = canEditPRTitle(pr.state)
-  // Tapping the state badge or the #number opens the PR on its host (GitHub/etc.)
-  // in the phone browser — pr.url is the canonical web URL.
+  // Badge, #number, and the flush-right external-link control all open pr.url
+  // (canonical host web URL) in the phone browser.
   const openPr = pr.url ? () => openMobilePrUrl(pr.url) : undefined
 
   return (

mobile/src/components/pr-sidebar/pr-auto-merge-availability.ts

@@ -15,6 +15,10 @@ function isConflicting(item: MobilePRAutoMergeAvailabilityInput): boolean {
   return item.mergeable === 'CONFLICTING' || item.mergeStateStatus === 'DIRTY'
 }
 
+function isUnstable(item: MobilePRAutoMergeAvailabilityInput): boolean {
+  return item.mergeStateStatus === 'UNSTABLE'
+}
+
 function hasReviewRequirement(item: MobilePRAutoMergeAvailabilityInput): boolean {
   return item.reviewDecision === 'REVIEW_REQUIRED' || item.reviewDecision === 'CHANGES_REQUESTED'
 }
@@ -27,7 +31,9 @@ function canMergeImmediately(item: MobilePRAutoMergeAvailabilityInput): boolean
 }
 
 function canRequestWhenReady(item: MobilePRAutoMergeAvailabilityInput): boolean {
-  if (item.state !== 'open' || isConflicting(item)) {
+  // Why: GitHub auto-merge waits on unmet requirements; UNSTABLE is rejected
+  // by the mutation rather than becoming a waitable auto-merge request.
+  if (item.state !== 'open' || isConflicting(item) || isUnstable(item)) {
     return false
   }
   if (item.mergeQueueRequired === true) {

mobile/src/files/MobileFileExplorerPanel.test.ts

@@ -0,0 +1,398 @@
+import { createElement } from 'react'
+import { act, create, type ReactTestRenderer } from 'react-test-renderer'
+import { afterEach, describe, expect, it, vi } from 'vitest'
+import { MobileFileExplorerPanel } from './MobileFileExplorerPanel'
+import type { MobileDirEntry } from './file-tree'
+import type { RpcResponse } from '../transport/types'
+
+type MockClient = {
+  sendRequest: ReturnType<typeof vi.fn>
+}
+
+const mockTransport = vi.hoisted(() => ({
+  client: null as MockClient | null,
+  connectionState: 'connected',
+  forceReconnect: vi.fn()
+}))
+
+vi.mock('react-native', async () => {
+  const React = await import('react')
+  return {
+    ActivityIndicator: 'ActivityIndicator',
+    FlatList: (props: {
+      data: unknown[]
+      keyExtractor: (item: unknown, index: number) => string
+      renderItem: (info: {
+        item: unknown
+        index: number
+        separators: Record<string, never>
+      }) => unknown
+    }) =>
+      React.createElement(
+        'FlatList',
+        props,
+        props.data.map((item, index) =>
+          React.createElement(
+            'FlatListItem',
+            { key: props.keyExtractor(item, index) },
+            props.renderItem({ item, index, separators: {} })
+          )
+        )
+      ),
+    Pressable: 'Pressable',
+    StyleSheet: {
+      create: (styles: unknown) => styles,
+      hairlineWidth: 1
+    },
+    Text: 'Text',
+    View: 'View'
+  }
+})
+
+vi.mock('react-native-safe-area-context', () => ({
+  SafeAreaView: 'SafeAreaView'
+}))
+
+vi.mock('expo-router', () => ({
+  useRouter: () => ({
+    back: vi.fn(),
+    push: vi.fn()
+  })
+}))
+
+vi.mock('lucide-react-native', () => ({
+  ChevronDown: 'ChevronDown',
+  ChevronLeft: 'ChevronLeft',
+  ChevronRight: 'ChevronRight',
+  File: 'File',
+  FileText: 'FileText',
+  Folder: 'Folder',
+  Image: 'Image',
+  X: 'X'
+}))
+
+vi.mock('../platform/haptics', () => ({
+  triggerSelection: vi.fn()
+}))
+
+vi.mock('../transport/client-context', () => ({
+  useForceReconnect: () => mockTransport.forceReconnect,
+  useHostClient: () => ({
+    client: mockTransport.client,
+    state: mockTransport.connectionState
+  })
+}))
+
+function suppressReactTestRendererDeprecationWarning(): () => void {
+  const originalConsoleError = console.error
+  const consoleErrorSpy = vi.spyOn(console, 'error').mockImplementation((...args) => {
+    const firstArg = args[0]
+    if (typeof firstArg === 'string' && firstArg.includes('react-test-renderer is deprecated')) {
+      return
+    }
+    originalConsoleError(...args)
+  })
+  return () => consoleErrorSpy.mockRestore()
+}
+
+function entry(name: string, isDirectory = false): MobileDirEntry {
+  return { name, isDirectory }
+}
+
+function ok(result: MobileDirEntry[]): RpcResponse {
+  return { id: 'response-id', ok: true, result, _meta: { runtimeId: 'runtime-id' } }
+}
+
+function createMockClient(entriesByPath: Record<string, MobileDirEntry[]>): MockClient {
+  return {
+    sendRequest: vi.fn(async (_method: string, params: { relativePath: string }) => {
+      return ok(entriesByPath[params.relativePath] ?? [])
+    })
+  }
+}
+
+async function renderExplorer(): Promise<ReactTestRenderer> {
+  let renderer: ReactTestRenderer | null = null
+  const restoreConsoleError = suppressReactTestRendererDeprecationWarning()
+  try {
+    await act(async () => {
+      renderer = create(
+        createElement(MobileFileExplorerPanel, {
+          hostId: 'host-a',
+          worktreeId: 'worktree-a',
+          name: 'Example Worktree',
+          embedded: true
+        })
+      )
+    })
+  } finally {
+    restoreConsoleError()
+  }
+  if (!renderer) {
+    throw new Error('MobileFileExplorerPanel did not render')
+  }
+  return renderer
+}
+
+async function updateExplorer(renderer: ReactTestRenderer): Promise<void> {
+  await act(async () => {
+    renderer.update(
+      createElement(MobileFileExplorerPanel, {
+        hostId: 'host-a',
+        worktreeId: 'worktree-a',
+        name: 'Example Worktree',
+        embedded: true
+      })
+    )
+  })
+}
+
+async function pressByLabel(
+  renderer: ReactTestRenderer,
+  accessibilityLabel: string
+): Promise<void> {
+  const pressable = renderer.root
+    .findAllByType('Pressable')
+    .find((node) => node.props.accessibilityLabel === accessibilityLabel)
+  if (!pressable) {
+    throw new Error(`Unable to find pressable: ${accessibilityLabel}`)
+  }
+  await act(async () => {
+    pressable.props.onPress()
+  })
+}
+
+function renderedText(renderer: ReactTestRenderer): string {
+  return renderer.root
+    .findAllByType('Text')
+    .flatMap((node) => node.props.children)
+    .join(' ')
+}
+
+describe('MobileFileExplorerPanel', () => {
+  afterEach(() => {
+    vi.restoreAllMocks()
+    mockTransport.client = null
+    mockTransport.connectionState = 'connected'
+    mockTransport.forceReconnect = vi.fn()
+  })
+
+  it('loads directories lazily and reuses cached children when reopened', async () => {
+    const client = createMockClient({
+      '': [entry('src', true), entry('README.md')],
+      src: [entry('app.ts'), entry('components', true)]
+    })
+    mockTransport.client = client
+
+    const renderer = await renderExplorer()
+
+    expect(client.sendRequest).toHaveBeenCalledTimes(1)
+    expect(client.sendRequest).toHaveBeenLastCalledWith('files.readDir', {
+      worktree: 'id:worktree-a',
+      relativePath: ''
+    })
+
+    await pressByLabel(renderer, 'Open folder src')
+    await vi.waitFor(() => expect(client.sendRequest).toHaveBeenCalledTimes(2))
+    expect(client.sendRequest).toHaveBeenLastCalledWith('files.readDir', {
+      worktree: 'id:worktree-a',
+      relativePath: 'src'
+    })
+    expect(renderedText(renderer)).toContain('app.ts')
+
+    await pressByLabel(renderer, 'Open folder src')
+    await pressByLabel(renderer, 'Open folder src')
+
+    expect(client.sendRequest).toHaveBeenCalledTimes(2)
+  })
+
+  it('keeps the loaded tree visible during a transient disconnect', async () => {
+    const client = createMockClient({
+      '': [entry('src', true), entry('README.md')]
+    })
+    mockTransport.client = client
+
+    const renderer = await renderExplorer()
+    expect(renderedText(renderer)).toContain('src')
+
+    mockTransport.client = null
+    mockTransport.connectionState = 'disconnected'
+    await updateExplorer(renderer)
+
+    expect(client.sendRequest).toHaveBeenCalledTimes(1)
+    expect(renderedText(renderer)).toContain('src')
+    expect(renderedText(renderer)).toContain('README.md')
+    expect(renderedText(renderer)).not.toContain('Waiting for desktop...')
+  })
+
+  it('refreshes the root in the background after reconnect without blanking the tree', async () => {
+    const client = createMockClient({
+      '': [entry('src', true), entry('README.md')]
+    })
+    mockTransport.client = client
+
+    const renderer = await renderExplorer()
+    expect(renderedText(renderer)).toContain('README.md')
+
+    mockTransport.client = null
+    mockTransport.connectionState = 'disconnected'
+    await updateExplorer(renderer)
+
+    let resolveReload: ((response: RpcResponse) => void) | undefined
+    const reconnectedClient: MockClient = {
+      sendRequest: vi.fn(
+        () =>
+          new Promise<RpcResponse>((resolve) => {
+            resolveReload = resolve
+          })
+      )
+    }
+    mockTransport.client = reconnectedClient
+    mockTransport.connectionState = 'connected'
+    await updateExplorer(renderer)
+
+    expect(reconnectedClient.sendRequest).toHaveBeenCalledTimes(1)
+    expect(renderedText(renderer)).toContain('src')
+    expect(renderedText(renderer)).toContain('README.md')
+
+    await act(async () => {
+      resolveReload?.(ok([entry('src', true), entry('README.md'), entry('CHANGELOG.md')]))
+    })
+    expect(renderedText(renderer)).toContain('CHANGELOG.md')
+  })
+
+  it('keeps the cached tree when a post-reconnect root refresh fails', async () => {
+    const client = createMockClient({
+      '': [entry('src', true), entry('README.md')]
+    })
+    mockTransport.client = client
+
+    const renderer = await renderExplorer()
+    expect(renderedText(renderer)).toContain('README.md')
+
+    mockTransport.client = null
+    mockTransport.connectionState = 'disconnected'
+    await updateExplorer(renderer)
+
+    const failingClient: MockClient = {
+      sendRequest: vi.fn(async () => {
+        throw new Error('refresh failed')
+      })
+    }
+    mockTransport.client = failingClient
+    mockTransport.connectionState = 'connected'
+    await updateExplorer(renderer)
+
+    expect(failingClient.sendRequest).toHaveBeenCalledTimes(1)
+    expect(renderedText(renderer)).toContain('src')
+    expect(renderedText(renderer)).toContain('README.md')
+    expect(renderedText(renderer)).not.toContain('refresh failed')
+  })
+
+  it('falls back to the capped files.list against desktops without files.readDir', async () => {
+    const legacyClient: MockClient = {
+      sendRequest: vi.fn(async (method: string): Promise<RpcResponse> => {
+        if (method === 'files.readDir') {
+          return {
+            id: 'response-id',
+            ok: false,
+            error: {
+              code: 'forbidden',
+              message: "Method 'files.readDir' is not available to mobile clients"
+            },
+            _meta: { runtimeId: 'runtime-id' }
+          }
+        }
+        return {
+          id: 'response-id',
+          ok: true,
+          result: {
+            files: [
+              { relativePath: 'src/app.ts', basename: 'app.ts', kind: 'text' },
+              { relativePath: 'README.md', basename: 'README.md', kind: 'text' }
+            ],
+            totalCount: 2,
+            truncated: false
+          },
+          _meta: { runtimeId: 'runtime-id' }
+        }
+      })
+    }
+    mockTransport.client = legacyClient
+
+    const renderer = await renderExplorer()
+
+    expect(legacyClient.sendRequest).toHaveBeenCalledWith('files.list', {
+      worktree: 'id:worktree-a'
+    })
+    expect(renderedText(renderer)).toContain('src')
+    expect(renderedText(renderer)).toContain('README.md')
+
+    await pressByLabel(renderer, 'Open folder src')
+    expect(renderedText(renderer)).toContain('app.ts')
+    // Every directory comes from the synthesized cache: one readDir attempt
+    // plus one files.list call total, no per-directory RPCs afterwards.
+    expect(legacyClient.sendRequest).toHaveBeenCalledTimes(2)
+  })
+
+  it('surfaces the legacy cap note when the files.list fallback is truncated', async () => {
+    const legacyClient: MockClient = {
+      sendRequest: vi.fn(async (method: string): Promise<RpcResponse> => {
+        if (method === 'files.readDir') {
+          return {
+            id: 'response-id',
+            ok: false,
+            error: { code: 'method_not_found', message: 'Unknown method' },
+            _meta: { runtimeId: 'runtime-id' }
+          }
+        }
+        return {
+          id: 'response-id',
+          ok: true,
+          result: {
+            files: [{ relativePath: 'README.md', basename: 'README.md', kind: 'text' }],
+            totalCount: 6000,
+            truncated: true
+          },
+          _meta: { runtimeId: 'runtime-id' }
+        }
+      })
+    }
+    mockTransport.client = legacyClient
+
+    const renderer = await renderExplorer()
+
+    expect(renderedText(renderer)).toContain('README.md')
+    expect(renderedText(renderer)).toContain('Showing first 5000')
+  })
+
+  it('reports the files.list failure when the fallback itself fails', async () => {
+    const legacyClient: MockClient = {
+      sendRequest: vi.fn(async (method: string): Promise<RpcResponse> => {
+        if (method === 'files.readDir') {
+          return {
+            id: 'response-id',
+            ok: false,
+            error: {
+              code: 'forbidden',
+              message: "Method 'files.readDir' is not available to mobile clients"
+            },
+            _meta: { runtimeId: 'runtime-id' }
+          }
+        }
+        return {
+          id: 'response-id',
+          ok: false,
+          error: { code: 'internal', message: 'legacy list failed' },
+          _meta: { runtimeId: 'runtime-id' }
+        }
+      })
+    }
+    mockTransport.client = legacyClient
+
+    const renderer = await renderExplorer()
+
+    expect(renderedText(renderer)).toContain('legacy list failed')
+    expect(renderedText(renderer)).not.toContain('not available to mobile clients')
+  })
+})

mobile/src/files/MobileFileExplorerPanel.tsx

@@ -1,4 +1,4 @@
-import { useCallback, useEffect, useMemo, useState } from 'react'
+import { useCallback, useEffect, useMemo, useRef, useState } from 'react'
 import {
   ActivityIndicator,
   FlatList,
@@ -9,34 +9,33 @@ import {
 } from 'react-native'
 import { SafeAreaView } from 'react-native-safe-area-context'
 import { useRouter } from 'expo-router'
-import {
-  ChevronDown,
-  ChevronLeft,
-  ChevronRight,
-  File,
-  FileText,
-  Folder,
-  Image as ImageIcon,
-  X
-} from 'lucide-react-native'
+import { ChevronLeft, X } from 'lucide-react-native'
 import { useHostClient, useForceReconnect } from '../transport/client-context'
 import { getWorktreeLabel } from '../session/worktree-label'
 import {
-  buildTree,
-  flattenTree,
-  isMarkdownPath,
-  type FilesListResult,
-  type MobileFileEntry,
-  type TreeNode
+  flattenDirectoryCache,
+  getDirectoryCacheState,
+  type DirectoryCache,
+  type FileExplorerRow,
+  type MobileDirEntry
 } from './file-tree'
 import type { RpcSuccess } from '../transport/types'
-import { triggerSelection } from '../platform/haptics'
-import { colors, spacing } from '../theme/mobile-theme'
-import { fileExplorerStyles as styles } from './mobile-file-explorer-styles'
+import { colors } from '../theme/mobile-theme'
 import {
-  canPreviewMobileFileRow,
-  navigateToMobileFilePreview
-} from './mobile-file-preview-navigation'
+  beginDirectoryLoad,
+  createDirectoryLoadRevisions,
+  isCurrentDirectoryLoad,
+  resetDirectoryLoadRevisions,
+  type DirectoryLoadRevisions
+} from './directory-load-revisions'
+import {
+  directoryCacheFromFileList,
+  isMobileMethodUnavailableError,
+  type LegacyFilesListResult
+} from './file-list-fallback'
+import { fileExplorerStyles as styles } from './mobile-file-explorer-styles'
+import { MobileFileExplorerRow } from './mobile-file-explorer-row'
+import { navigateToMobileFilePreview } from './mobile-file-preview-navigation'
 
 export function MobileFileExplorerPanel(props: {
   hostId: string
@@ -49,58 +48,217 @@ export function MobileFileExplorerPanel(props: {
   const router = useRouter()
   const { client, state: connState } = useHostClient(hostId)
   const forceReconnect = useForceReconnect()
-  const [files, setFiles] = useState<MobileFileEntry[]>([])
+  const scopeRef = useRef('')
+  const scope = `${hostId}:${worktreeId}`
+  scopeRef.current = scope
+  const directoryLoadRevisionsRef = useRef<DirectoryLoadRevisions>(createDirectoryLoadRevisions())
+  const pendingDirectoryRetriesRef = useRef<Set<string>>(new Set())
+  const directoryCacheRef = useRef<DirectoryCache>({})
+  const [directoryCache, setDirectoryCache] = useState<DirectoryCache>({})
   const [expanded, setExpanded] = useState<Set<string>>(() => new Set())
   const [loading, setLoading] = useState(true)
   const [error, setError] = useState<string | null>(null)
-  const [truncated, setTruncated] = useState(false)
+  const [legacyListTruncated, setLegacyListTruncated] = useState(false)
   const worktreeLabel = getWorktreeLabel(name, worktreeId)
 
-  const loadFiles = useCallback(async () => {
-    if (!client || connState !== 'connected') {
-      setLoading(false)
-      setError(connState === 'connected' ? 'Connecting to desktop...' : 'Waiting for desktop...')
-      return
-    }
-    setLoading(true)
-    setError(null)
-    try {
-      const response = await client.sendRequest('files.list', { worktree: `id:${worktreeId}` })
-      if (!response.ok) {
-        throw new Error(response.error?.message || 'Unable to load files')
+  const loadDirectory = useCallback(
+    async (relativePath: string) => {
+      const scope = scopeRef.current
+      const loadToken = beginDirectoryLoad(directoryLoadRevisionsRef.current, scope, relativePath)
+      const rootLoad = relativePath === ''
+
+      if (!client || connState !== 'connected') {
+        const message =
+          connState === 'connected' ? 'Connecting to desktop...' : 'Waiting for desktop...'
+        if (rootLoad) {
+          const hasLoadedRoot =
+            (getDirectoryCacheState(directoryCacheRef.current, '')?.entries.length ?? 0) > 0
+          setLoading(false)
+          // Why: transient reconnects should not blank an already browsable tree.
+          setError(hasLoadedRoot ? null : message)
+        } else {
+          setDirectoryCache((prev) => ({
+            ...prev,
+            [relativePath]: {
+              entries: getDirectoryCacheState(prev, relativePath)?.entries ?? [],
+              error: message
+            }
+          }))
+        }
+        return
       }
-      const result = (response as RpcSuccess).result as FilesListResult
-      setFiles(result.files)
-      setTruncated(result.truncated)
-    } catch (err) {
-      setError(err instanceof Error ? err.message : 'Unable to load files')
-    } finally {
-      setLoading(false)
-    }
-  }, [client, connState, worktreeId])
+
+      const hadLoadedRoot =
+        rootLoad && (getDirectoryCacheState(directoryCacheRef.current, '')?.entries.length ?? 0) > 0
+      if (rootLoad) {
+        // Why: a reconnect refresh must not blank an already browsable tree —
+        // the full-screen spinner unmounts the list and resets scroll.
+        if (!hadLoadedRoot) {
+          setLoading(true)
+        }
+        setError(null)
+      }
+      setDirectoryCache((prev) => ({
+        ...prev,
+        [relativePath]: {
+          entries: getDirectoryCacheState(prev, relativePath)?.entries ?? [],
+          loading: true
+        }
+      }))
+
+      try {
+        const response = await client.sendRequest('files.readDir', {
+          worktree: `id:${worktreeId}`,
+          relativePath
+        })
+        if (!response.ok) {
+          // Why: desktops that predate the files.readDir mobile allowlist
+          // entry still serve the capped files.list; fall back so the Files
+          // tab keeps working until the desktop updates.
+          if (
+            rootLoad &&
+            isMobileMethodUnavailableError(response.error?.code, response.error?.message)
+          ) {
+            const legacy = await client.sendRequest('files.list', {
+              worktree: `id:${worktreeId}`
+            })
+            if (legacy.ok) {
+              if (
+                !isCurrentDirectoryLoad(
+                  directoryLoadRevisionsRef.current,
+                  scopeRef.current,
+                  loadToken
+                )
+              ) {
+                return
+              }
+              const legacyResult = (legacy as RpcSuccess).result as LegacyFilesListResult
+              setDirectoryCache(directoryCacheFromFileList(legacyResult.files))
+              // Why: the capped list silently omits files past the cap — keep
+              // the legacy explorer's "Showing first 5000" note.
+              setLegacyListTruncated(legacyResult.truncated)
+              return
+            }
+            throw new Error(
+              legacy.error?.message || response.error?.message || 'Unable to load files'
+            )
+          }
+          throw new Error(response.error?.message || 'Unable to load files')
+        }
+        if (
+          !isCurrentDirectoryLoad(directoryLoadRevisionsRef.current, scopeRef.current, loadToken)
+        ) {
+          return
+        }
+        const entries = (response as RpcSuccess).result as MobileDirEntry[]
+        if (rootLoad) {
+          setLegacyListTruncated(false)
+        }
+        setDirectoryCache((prev) => ({
+          ...prev,
+          [relativePath]: { entries }
+        }))
+      } catch (err) {
+        if (
+          !isCurrentDirectoryLoad(directoryLoadRevisionsRef.current, scopeRef.current, loadToken)
+        ) {
+          return
+        }
+        const message = err instanceof Error ? err.message : 'Unable to load files'
+        if (rootLoad) {
+          // Why: a failed background refresh keeps the cached tree browsable;
+          // only a cold load surfaces the full-screen error.
+          setError(hadLoadedRoot ? null : message)
+        } else {
+          setDirectoryCache((prev) => ({
+            ...prev,
+            [relativePath]: {
+              entries: getDirectoryCacheState(prev, relativePath)?.entries ?? [],
+              error: message
+            }
+          }))
+        }
+      } finally {
+        if (
+          rootLoad &&
+          isCurrentDirectoryLoad(directoryLoadRevisionsRef.current, scopeRef.current, loadToken)
+        ) {
+          setLoading(false)
+        }
+      }
+    },
+    [client, connState, worktreeId]
+  )
 
   useEffect(() => {
-    void loadFiles()
-  }, [loadFiles])
+    scopeRef.current = scope
+    resetDirectoryLoadRevisions(directoryLoadRevisionsRef.current)
+    pendingDirectoryRetriesRef.current.clear()
+    directoryCacheRef.current = {}
+    setDirectoryCache({})
+    setExpanded(new Set())
+    setLoading(true)
+    setError(null)
+    setLegacyListTruncated(false)
+  }, [scope])
 
-  const rows = useMemo(() => flattenTree(buildTree(files), expanded), [expanded, files])
+  useEffect(() => {
+    directoryCacheRef.current = directoryCache
+  }, [directoryCache])
 
-  const toggleDirectory = useCallback((relativePath: string) => {
-    triggerSelection()
-    setExpanded((prev) => {
-      const next = new Set(prev)
-      if (next.has(relativePath)) {
-        next.delete(relativePath)
-      } else {
-        next.add(relativePath)
+  useEffect(() => {
+    void loadDirectory('')
+  }, [hostId, loadDirectory])
+
+  useEffect(() => {
+    if (connState !== 'connected' || pendingDirectoryRetriesRef.current.size === 0) {
+      return
+    }
+    const pending = [...pendingDirectoryRetriesRef.current]
+    pendingDirectoryRetriesRef.current.clear()
+    for (const relativePath of pending) {
+      void loadDirectory(relativePath)
+    }
+  }, [connState, loadDirectory])
+
+  const rows = useMemo(
+    () => flattenDirectoryCache(directoryCache, expanded),
+    [directoryCache, expanded]
+  )
+
+  const toggleDirectory = useCallback(
+    (relativePath: string) => {
+      setExpanded((prev) => {
+        const next = new Set(prev)
+        if (next.has(relativePath)) {
+          next.delete(relativePath)
+        } else {
+          next.add(relativePath)
+        }
+        return next
+      })
+      const state = getDirectoryCacheState(directoryCache, relativePath)
+      if (!expanded.has(relativePath) && !state?.loading && (!state?.entries || state.error)) {
+        void loadDirectory(relativePath)
       }
-      return next
-    })
-  }, [])
+    },
+    [directoryCache, expanded, loadDirectory]
+  )
+
+  const retryDirectory = useCallback(
+    (relativePath: string) => {
+      if (connState !== 'connected' && hostId) {
+        pendingDirectoryRetriesRef.current.add(relativePath)
+        void forceReconnect(hostId)
+        return
+      }
+      void loadDirectory(relativePath)
+    },
+    [connState, forceReconnect, hostId, loadDirectory]
+  )
 
   const previewFile = useCallback(
     (relativePath: string, displayName: string) => {
-      triggerSelection()
       navigateToMobileFilePreview(
         router,
         {
@@ -116,66 +274,15 @@ export function MobileFileExplorerPanel(props: {
     [embedded, hostId, name, onRequestClose, router, worktreeId]
   )
 
-  const renderItem: ListRenderItem<TreeNode> = ({ item }) => {
-    const isDirectory = item.kind === 'directory'
-    const isExpanded = expanded.has(item.relativePath)
-    // Images render in the mobile viewer (via files.readPreview), so a binary
-    // image is openable; only non-previewable binaries are unavailable.
-    const previewable =
-      item.kind !== 'directory' &&
-      canPreviewMobileFileRow({ kind: item.kind, relativePath: item.relativePath })
-    const isImage = item.kind === 'binary' && previewable
-    const disabled = item.kind === 'binary' && !previewable
-    const markdown = item.kind === 'text' && isMarkdownPath(item.relativePath)
+  const renderItem: ListRenderItem<FileExplorerRow> = ({ item }) => {
     return (
-      <Pressable
-        style={({ pressed }) => [
-          styles.row,
-          { paddingLeft: spacing.lg + item.depth * 18 },
-          pressed && !disabled && styles.rowPressed,
-          disabled && styles.rowDisabled
-        ]}
-        disabled={disabled}
-        onPress={() => {
-          if (isDirectory) {
-            toggleDirectory(item.relativePath)
-          } else if (!disabled) {
-            previewFile(item.relativePath, item.name)
-          }
-        }}
-        accessibilityLabel={
-          isDirectory
-            ? `Open folder ${item.name}`
-            : disabled
-              ? `${item.name} unavailable on mobile`
-              : `Preview file ${item.name}`
-        }
-      >
-        {isDirectory ? (
-          isExpanded ? (
-            <ChevronDown size={16} color={colors.textSecondary} />
-          ) : (
-            <ChevronRight size={16} color={colors.textSecondary} />
-          )
-        ) : (
-          <View style={styles.chevronSpacer} />
-        )}
-        {isDirectory ? (
-          <Folder size={17} color={colors.textSecondary} />
-        ) : markdown ? (
-          <FileText size={17} color={disabled ? colors.textMuted : colors.textSecondary} />
-        ) : isImage ? (
-          <ImageIcon size={17} color={colors.textSecondary} />
-        ) : (
-          <File size={17} color={disabled ? colors.textMuted : colors.textSecondary} />
-        )}
-        <View style={styles.rowTextBlock}>
-          <Text style={[styles.rowTitle, disabled && styles.rowTitleDisabled]} numberOfLines={1}>
-            {item.name}
-          </Text>
-          {disabled ? <Text style={styles.rowMeta}>Unavailable on mobile</Text> : null}
-        </View>
-      </Pressable>
+      <MobileFileExplorerRow
+        item={item}
+        expanded={expanded}
+        onPreviewFile={previewFile}
+        onRetryDirectory={retryDirectory}
+        onToggleDirectory={toggleDirectory}
+      />
     )
   }
 
@@ -206,7 +313,7 @@ export function MobileFileExplorerPanel(props: {
         </Text>
         <Text style={styles.meta} numberOfLines={1}>
           {worktreeLabel}
-          {truncated ? ' - Showing first 5000' : ''}
+          {legacyListTruncated ? ' - Showing first 5000' : ''}
         </Text>
       </View>
     </View>
@@ -220,12 +327,12 @@ export function MobileFileExplorerPanel(props: {
     <View style={styles.state}>
       <Text style={styles.errorText}>{error}</Text>
       {/* Why: while disconnected, re-sending the request is useless — revive
-          the parked transport instead (issue #5049); loadFiles re-runs via
+          the parked transport instead (issue #5049); loadDirectory re-runs via
           its effect once the new client connects. */}
       <Pressable
         style={styles.retryButton}
         onPress={() =>
-          connState !== 'connected' && hostId ? void forceReconnect(hostId) : void loadFiles()
+          connState !== 'connected' && hostId ? void forceReconnect(hostId) : void loadDirectory('')
         }
       >
         <Text style={styles.retryText}>Retry</Text>

mobile/src/files/directory-load-revisions.test.ts

@@ -0,0 +1,44 @@
+import { describe, expect, it } from 'vitest'
+import {
+  beginDirectoryLoad,
+  createDirectoryLoadRevisions,
+  isCurrentDirectoryLoad,
+  resetDirectoryLoadRevisions,
+  type DirectoryLoadRevisions
+} from './directory-load-revisions'
+
+describe('directory-load-revisions', () => {
+  it('ignores stale duplicate loads for the same directory and scope', () => {
+    const revisions: DirectoryLoadRevisions = createDirectoryLoadRevisions()
+    const older = beginDirectoryLoad(revisions, 'host-a:worktree-a', 'src')
+    const newer = beginDirectoryLoad(revisions, 'host-a:worktree-a', 'src')
+
+    expect(isCurrentDirectoryLoad(revisions, 'host-a:worktree-a', newer)).toBe(true)
+    expect(isCurrentDirectoryLoad(revisions, 'host-a:worktree-a', older)).toBe(false)
+  })
+
+  it('ignores loads from an old host/worktree scope', () => {
+    const revisions: DirectoryLoadRevisions = createDirectoryLoadRevisions()
+    const oldScopeLoad = beginDirectoryLoad(revisions, 'host-a:worktree-a', '')
+
+    expect(isCurrentDirectoryLoad(revisions, 'host-b:worktree-b', oldScopeLoad)).toBe(false)
+  })
+
+  it('ignores loads from an old reset generation in the same scope', () => {
+    const revisions: DirectoryLoadRevisions = createDirectoryLoadRevisions()
+    const oldLoad = beginDirectoryLoad(revisions, 'host-a:worktree-a', '')
+
+    resetDirectoryLoadRevisions(revisions)
+    const newLoad = beginDirectoryLoad(revisions, 'host-a:worktree-a', '')
+
+    expect(isCurrentDirectoryLoad(revisions, 'host-a:worktree-a', newLoad)).toBe(true)
+    expect(isCurrentDirectoryLoad(revisions, 'host-a:worktree-a', oldLoad)).toBe(false)
+  })
+
+  it('tracks directory names that overlap object prototype keys', () => {
+    const revisions: DirectoryLoadRevisions = createDirectoryLoadRevisions()
+    const load = beginDirectoryLoad(revisions, 'host-a:worktree-a', '__proto__')
+
+    expect(isCurrentDirectoryLoad(revisions, 'host-a:worktree-a', load)).toBe(true)
+  })
+})

mobile/src/files/directory-load-revisions.ts

@@ -0,0 +1,42 @@
+export type DirectoryLoadRevisions = {
+  generation: number
+  revisionsByPath: Map<string, number>
+}
+
+export type DirectoryLoadToken = {
+  generation: number
+  relativePath: string
+  revision: number
+  scope: string
+}
+
+export function createDirectoryLoadRevisions(): DirectoryLoadRevisions {
+  return { generation: 0, revisionsByPath: new Map() }
+}
+
+export function resetDirectoryLoadRevisions(revisions: DirectoryLoadRevisions): void {
+  revisions.generation += 1
+  revisions.revisionsByPath.clear()
+}
+
+export function beginDirectoryLoad(
+  revisions: DirectoryLoadRevisions,
+  scope: string,
+  relativePath: string
+): DirectoryLoadToken {
+  const revision = (revisions.revisionsByPath.get(relativePath) ?? 0) + 1
+  revisions.revisionsByPath.set(relativePath, revision)
+  return { generation: revisions.generation, relativePath, revision, scope }
+}
+
+export function isCurrentDirectoryLoad(
+  revisions: DirectoryLoadRevisions,
+  currentScope: string,
+  token: DirectoryLoadToken
+): boolean {
+  return (
+    currentScope === token.scope &&
+    revisions.generation === token.generation &&
+    revisions.revisionsByPath.get(token.relativePath) === token.revision
+  )
+}

mobile/src/files/file-list-fallback.test.ts

@@ -0,0 +1,65 @@
+import { describe, expect, it } from 'vitest'
+import { directoryCacheFromFileList, isMobileMethodUnavailableError } from './file-list-fallback'
+import { getDirectoryCacheState } from './file-tree'
+
+describe('isMobileMethodUnavailableError', () => {
+  it('detects old-desktop allowlist and missing-method failures', () => {
+    expect(isMobileMethodUnavailableError('forbidden', undefined)).toBe(true)
+    expect(isMobileMethodUnavailableError('method_not_found', undefined)).toBe(true)
+    expect(
+      isMobileMethodUnavailableError(
+        'some_code',
+        "Method 'files.readDir' is not available to mobile clients"
+      )
+    ).toBe(true)
+    expect(isMobileMethodUnavailableError('internal', 'boom')).toBe(false)
+    expect(isMobileMethodUnavailableError(undefined, undefined)).toBe(false)
+  })
+})
+
+describe('directoryCacheFromFileList', () => {
+  it('synthesizes every ancestor directory from flat paths', () => {
+    const cache = directoryCacheFromFileList([
+      { relativePath: 'src/lib/util.ts', basename: 'util.ts', kind: 'text' },
+      { relativePath: 'src/app.ts', basename: 'app.ts', kind: 'text' },
+      { relativePath: 'README.md', basename: 'README.md', kind: 'text' }
+    ])
+    expect(cache['']?.entries).toEqual(
+      expect.arrayContaining([
+        { name: 'src', isDirectory: true },
+        { name: 'README.md', isDirectory: false }
+      ])
+    )
+    expect(cache['src']?.entries).toEqual(
+      expect.arrayContaining([
+        { name: 'lib', isDirectory: true },
+        { name: 'app.ts', isDirectory: false }
+      ])
+    )
+    expect(cache['src/lib']?.entries).toEqual([{ name: 'util.ts', isDirectory: false }])
+  })
+
+  it('keeps a name a directory when it appears as both file and dir prefix', () => {
+    const cache = directoryCacheFromFileList([
+      { relativePath: 'src', basename: 'src', kind: 'text' },
+      { relativePath: 'src/app.ts', basename: 'app.ts', kind: 'text' }
+    ])
+    expect(cache['']?.entries).toEqual([{ name: 'src', isDirectory: true }])
+  })
+
+  it('returns an empty root for an empty list', () => {
+    const cache = directoryCacheFromFileList([])
+    expect(cache['']?.entries).toEqual([])
+  })
+
+  it('stores a __proto__ directory as an own key instead of mutating the prototype', () => {
+    const cache = directoryCacheFromFileList([
+      { relativePath: '__proto__/pollute.js', basename: 'pollute.js', kind: 'text' }
+    ])
+    expect(Object.getPrototypeOf(cache)).toBe(Object.prototype)
+    expect(cache['']?.entries).toEqual([{ name: '__proto__', isDirectory: true }])
+    expect(getDirectoryCacheState(cache, '__proto__')?.entries).toEqual([
+      { name: 'pollute.js', isDirectory: false }
+    ])
+  })
+})

mobile/src/files/file-list-fallback.ts

@@ -0,0 +1,73 @@
+// Fallback for desktops that predate files.readDir in the mobile RPC
+// allowlist: synthesize the lazy directory cache from the flat, capped
+// files.list result so the Files tab stays browsable against old desktops.
+import type { DirectoryCache, MobileDirEntry } from './file-tree'
+
+export type LegacyMobileFileEntry = {
+  relativePath: string
+  basename: string
+  kind: 'text' | 'binary'
+}
+
+export type LegacyFilesListResult = {
+  files: LegacyMobileFileEntry[]
+  totalCount: number
+  truncated: boolean
+}
+
+// Same detection shape as isMobileGitUnavailable in mobile-git-status.ts:
+// 'forbidden' = method exists but is not mobile-allowlisted on the old
+// desktop; 'method_not_found' = desktop predates the method entirely.
+export function isMobileMethodUnavailableError(
+  code: string | undefined,
+  message: string | undefined
+): boolean {
+  return (
+    code === 'forbidden' ||
+    code === 'method_not_found' ||
+    message?.includes('not available to mobile clients') === true
+  )
+}
+
+export function directoryCacheFromFileList(files: LegacyMobileFileEntry[]): DirectoryCache {
+  const childrenByDir = new Map<string, Map<string, boolean>>()
+  const ensureDir = (path: string): Map<string, boolean> => {
+    let children = childrenByDir.get(path)
+    if (!children) {
+      children = new Map()
+      childrenByDir.set(path, children)
+    }
+    return children
+  }
+  ensureDir('')
+  for (const file of files) {
+    const parts = file.relativePath.split('/').filter(Boolean)
+    let parentPath = ''
+    parts.forEach((name, index) => {
+      const isDirectory = index < parts.length - 1
+      const children = ensureDir(parentPath)
+      children.set(name, children.get(name) === true || isDirectory)
+      parentPath = parentPath ? `${parentPath}/${name}` : name
+      if (isDirectory) {
+        ensureDir(parentPath)
+      }
+    })
+  }
+  // Why: plain `cache[path] = ...` with a '__proto__' path segment mutates the
+  // object's prototype instead of storing the directory; fromEntries always
+  // creates own keys.
+  return Object.fromEntries(
+    Array.from(childrenByDir, ([path, children]) => [
+      path,
+      {
+        entries: Array.from(
+          children,
+          ([name, isDirectory]): MobileDirEntry => ({
+            name,
+            isDirectory
+          })
+        )
+      }
+    ])
+  )
+}

mobile/src/files/file-tree.test.ts

@@ -1,32 +1,109 @@
 import { describe, expect, it } from 'vitest'
-import { buildTree, flattenTree, isMarkdownPath, type MobileFileEntry } from './file-tree'
+import {
+  flattenDirectoryCache,
+  getMobileFileKind,
+  isMarkdownPath,
+  shouldIncludeMobileFileExplorerEntry,
+  type DirectoryCache,
+  type MobileDirEntry
+} from './file-tree'
 
-function entry(relativePath: string, kind: 'text' | 'binary' = 'text'): MobileFileEntry {
-  return { relativePath, basename: relativePath.split('/').pop() ?? relativePath, kind }
+function entry(name: string, isDirectory = false, isSymlink = false): MobileDirEntry {
+  return { name, isDirectory, isSymlink }
 }
 
 describe('file-tree', () => {
-  it('nests files under their directories', () => {
-    const root = buildTree([entry('src/app.ts'), entry('src/lib/util.ts'), entry('readme.md')])
-    expect(root.files.map((f) => f.relativePath)).toEqual(['readme.md'])
-    expect(root.directories.get('src')?.directories.get('lib')?.files[0]?.relativePath).toBe(
-      'src/lib/util.ts'
-    )
+  it('flattens cached directories lazily with directories before files', () => {
+    const cache: DirectoryCache = {
+      '': { entries: [entry('zeta.txt'), entry('src', true), entry('readme.md')] },
+      src: { entries: [entry('app.ts'), entry('lib', true)] },
+      'src/lib': { entries: [entry('util.ts')] }
+    }
+
+    const collapsed = flattenDirectoryCache(cache, new Set())
+    expect(collapsed.map((row) => row.id)).toEqual(['dir:src', 'file:readme.md', 'file:zeta.txt'])
+
+    const expanded = flattenDirectoryCache(cache, new Set(['src', 'src/lib']))
+    expect(expanded.map((row) => row.id)).toEqual([
+      'dir:src',
+      'dir:src/lib',
+      'file:src/lib/util.ts',
+      'file:src/app.ts',
+      'file:readme.md',
+      'file:zeta.txt'
+    ])
   })
 
-  it('flattens with directories before files and only expands open dirs', () => {
-    const root = buildTree([entry('src/app.ts'), entry('zeta.txt')])
-    const collapsed = flattenTree(root, new Set())
-    expect(collapsed.map((r) => r.id)).toEqual(['dir:src', 'file:zeta.txt'])
+  it('mirrors desktop default browse exclusions while keeping dotfiles visible', () => {
+    const cache: DirectoryCache = {
+      '': {
+        entries: [
+          entry('.git', true),
+          entry('.env'),
+          entry('node_modules', true),
+          entry('src', true)
+        ]
+      }
+    }
 
-    const expanded = flattenTree(root, new Set(['src']))
-    expect(expanded.map((r) => r.id)).toEqual(['dir:src', 'file:src/app.ts', 'file:zeta.txt'])
+    expect(flattenDirectoryCache(cache, new Set()).map((row) => row.id)).toEqual([
+      'dir:src',
+      'file:.env'
+    ])
+    expect(shouldIncludeMobileFileExplorerEntry(entry('.config', true))).toBe(true)
   })
 
-  it('preserves the binary kind on flattened rows', () => {
-    const root = buildTree([entry('assets/logo.png', 'binary')])
-    const rows = flattenTree(root, new Set(['assets']))
-    expect(rows.find((r) => r.id === 'file:assets/logo.png')?.kind).toBe('binary')
+  it('renders inline loading and error rows under expanded directories', () => {
+    const cache: DirectoryCache = {
+      '': { entries: [entry('loading-dir', true), entry('error-dir', true)] },
+      'loading-dir': { entries: [], loading: true },
+      'error-dir': { entries: [], error: 'permission denied' }
+    }
+
+    const rows = flattenDirectoryCache(cache, new Set(['loading-dir', 'error-dir']))
+    expect(rows.map((row) => row.id)).toEqual([
+      'dir:error-dir',
+      'error:error-dir',
+      'dir:loading-dir',
+      'loading:loading-dir'
+    ])
+    expect(rows.find((row) => row.id === 'error:error-dir')).toMatchObject({
+      kind: 'error',
+      message: 'permission denied'
+    })
+  })
+
+  it('preserves symlink metadata for user-initiated activation', () => {
+    const cache: DirectoryCache = {
+      '': { entries: [entry('linked-docs', false, true)] }
+    }
+
+    expect(flattenDirectoryCache(cache, new Set())).toEqual([
+      {
+        id: 'file:linked-docs',
+        name: 'linked-docs',
+        relativePath: 'linked-docs',
+        depth: 0,
+        kind: 'text',
+        isSymlink: true
+      }
+    ])
+  })
+
+  it('does not treat inherited object keys as loaded directories', () => {
+    const cache: DirectoryCache = {
+      '': { entries: [entry('constructor', true), entry('__proto__', true)] },
+      ['__proto__']: { entries: [entry('file.txt')] }
+    }
+
+    expect(
+      flattenDirectoryCache(cache, new Set(['constructor', '__proto__'])).map((row) => row.id)
+    ).toEqual(['dir:__proto__', 'file:__proto__/file.txt', 'dir:constructor'])
+  })
+
+  it('classifies binary file paths for existing mobile preview behavior', () => {
+    expect(getMobileFileKind('assets/logo.png')).toBe('binary')
+    expect(getMobileFileKind('src/app.ts')).toBe('text')
   })
 
   it('detects markdown paths', () => {

mobile/src/files/file-tree.ts

@@ -1,91 +1,152 @@
-// Pure tree model for the mobile file explorer: turns the flat files.list
-// result into a nested directory structure and flattens it into renderable
-// rows. Kept out of the screen component so the screen stays under its line cap.
+// Pure tree projection for the mobile file explorer. Mobile mirrors desktop
+// browse semantics by flattening cached files.readDir results as folders open.
 
-export type MobileFileEntry = {
-  relativePath: string
-  basename: string
-  kind: 'text' | 'binary'
+export type MobileDirEntry = {
+  name: string
+  isDirectory: boolean
+  isSymlink?: boolean
 }
 
-export type FilesListResult = {
-  files: MobileFileEntry[]
-  totalCount: number
-  truncated: boolean
-}
+export type MobileFileKind = 'text' | 'binary'
 
 export type TreeNode = {
   id: string
   name: string
   relativePath: string
   depth: number
-  kind: 'directory' | 'text' | 'binary'
+  kind: 'directory' | MobileFileKind
+  isSymlink?: boolean
 }
 
-export type DirectoryNode = {
-  name: string
+export type DirectoryState = {
+  entries: MobileDirEntry[]
+  loading?: boolean
+  error?: string
+}
+
+export type DirectoryCache = Record<string, DirectoryState | undefined>
+
+export type InlineStatusNode = {
+  id: string
   relativePath: string
-  directories: Map<string, DirectoryNode>
-  files: MobileFileEntry[]
+  depth: number
+  kind: 'loading' | 'error'
+  message?: string
 }
 
-function createDirectoryNode(name: string, relativePath: string): DirectoryNode {
-  return { name, relativePath, directories: new Map(), files: [] }
-}
+export type FileExplorerRow = TreeNode | InlineStatusNode
 
-export function buildTree(files: MobileFileEntry[]): DirectoryNode {
-  const root = createDirectoryNode('', '')
-  for (const file of files) {
-    const parts = file.relativePath.split('/').filter(Boolean)
-    let current = root
-    for (let index = 0; index < parts.length - 1; index += 1) {
-      const name = parts[index]!
-      const relativePath = parts.slice(0, index + 1).join('/')
-      let child = current.directories.get(name)
-      if (!child) {
-        child = createDirectoryNode(name, relativePath)
-        current.directories.set(name, child)
-      }
-      current = child
-    }
-    current.files.push(file)
-  }
-  return root
-}
+const DESKTOP_EXCLUDED_NAMES = new Set(['.git', 'node_modules'])
+const BINARY_EXTENSIONS = new Set([
+  '.avif',
+  '.bmp',
+  '.gif',
+  '.heic',
+  '.ico',
+  '.jpeg',
+  '.jpg',
+  '.mov',
+  '.mp3',
+  '.mp4',
+  '.pdf',
+  '.png',
+  '.webp',
+  '.zip'
+])
 
-export function flattenTree(root: DirectoryNode, expanded: ReadonlySet<string>): TreeNode[] {
-  const rows: TreeNode[] = []
-  const visit = (directory: DirectoryNode, depth: number): void => {
-    const dirs = Array.from(directory.directories.values()).sort((a, b) =>
-      a.name.localeCompare(b.name)
-    )
-    for (const child of dirs) {
-      rows.push({
-        id: `dir:${child.relativePath}`,
-        name: child.name,
-        relativePath: child.relativePath,
-        depth,
-        kind: 'directory'
-      })
-      if (expanded.has(child.relativePath)) {
-        visit(child, depth + 1)
-      }
-    }
-    const files = [...directory.files].sort((a, b) => a.basename.localeCompare(b.basename))
-    for (const file of files) {
-      rows.push({
-        id: `file:${file.relativePath}`,
-        name: file.basename,
-        relativePath: file.relativePath,
-        depth,
-        kind: file.kind
-      })
-    }
-  }
-  visit(root, 0)
+export function flattenDirectoryCache(
+  cache: DirectoryCache,
+  expanded: ReadonlySet<string>
+): FileExplorerRow[] {
+  const rows: FileExplorerRow[] = []
+  visitDirectory('', 0, cache, expanded, rows)
   return rows
 }
 
+function visitDirectory(
+  relativePath: string,
+  depth: number,
+  cache: DirectoryCache,
+  expanded: ReadonlySet<string>,
+  rows: FileExplorerRow[]
+): void {
+  const state = getDirectoryCacheState(cache, relativePath)
+  const entries = state?.entries ?? []
+  const visibleEntries = entries
+    .filter(shouldIncludeMobileFileExplorerEntry)
+    .sort(compareDirectoryEntries)
+
+  for (const entry of visibleEntries) {
+    const childPath = joinRelativePath(relativePath, entry.name)
+    rows.push(toTreeNode(entry, childPath, depth))
+    if (entry.isDirectory && expanded.has(childPath)) {
+      const childState = getDirectoryCacheState(cache, childPath)
+      if (childState?.loading) {
+        rows.push({
+          id: `loading:${childPath}`,
+          relativePath: childPath,
+          depth: depth + 1,
+          kind: 'loading'
+        })
+      } else if (childState?.error) {
+        rows.push({
+          id: `error:${childPath}`,
+          relativePath: childPath,
+          depth: depth + 1,
+          kind: 'error',
+          message: childState.error
+        })
+      } else {
+        visitDirectory(childPath, depth + 1, cache, expanded, rows)
+      }
+    }
+  }
+}
+
+function toTreeNode(entry: MobileDirEntry, relativePath: string, depth: number): TreeNode {
+  return {
+    id: `${entry.isDirectory ? 'dir' : 'file'}:${relativePath}`,
+    name: entry.name,
+    relativePath,
+    depth,
+    kind: entry.isDirectory ? 'directory' : getMobileFileKind(relativePath),
+    isSymlink: entry.isSymlink
+  }
+}
+
+function compareDirectoryEntries(a: MobileDirEntry, b: MobileDirEntry): number {
+  if (a.isDirectory !== b.isDirectory) {
+    return a.isDirectory ? -1 : 1
+  }
+  return a.name.localeCompare(b.name)
+}
+
+export function shouldIncludeMobileFileExplorerEntry(entry: MobileDirEntry): boolean {
+  return !DESKTOP_EXCLUDED_NAMES.has(entry.name)
+}
+
+export function getDirectoryCacheState(
+  cache: DirectoryCache,
+  relativePath: string
+): DirectoryState | undefined {
+  // Why: repository paths are arbitrary object keys; inherited keys like
+  // "constructor" must not masquerade as loaded directory state.
+  return Object.prototype.hasOwnProperty.call(cache, relativePath) ? cache[relativePath] : undefined
+}
+
+export function joinRelativePath(parentPath: string, name: string): string {
+  return parentPath ? `${parentPath}/${name}` : name
+}
+
+export function getMobileFileKind(relativePath: string): MobileFileKind {
+  const basename = relativePath.split('/').pop() ?? relativePath
+  const dotIndex = basename.lastIndexOf('.')
+  if (dotIndex <= 0) {
+    return 'text'
+  }
+  return BINARY_EXTENSIONS.has(basename.slice(dotIndex).toLowerCase()) ? 'binary' : 'text'
+}
+
 export function isMarkdownPath(relativePath: string): boolean {
   return /\.(md|mdx|markdown)$/i.test(relativePath)
 }

mobile/src/files/mobile-file-explorer-row.tsx

@@ -0,0 +1,145 @@
+import { ActivityIndicator, Pressable, Text, View } from 'react-native'
+import {
+  ChevronDown,
+  ChevronRight,
+  File,
+  FileText,
+  Folder,
+  Image as ImageIcon
+} from 'lucide-react-native'
+import { triggerSelection } from '../platform/haptics'
+import { colors, spacing } from '../theme/mobile-theme'
+import { type FileExplorerRow, isMarkdownPath, type TreeNode } from './file-tree'
+import { fileExplorerStyles as styles } from './mobile-file-explorer-styles'
+import { canPreviewMobileFileRow } from './mobile-file-preview-navigation'
+
+type Props = {
+  item: FileExplorerRow
+  expanded: ReadonlySet<string>
+  onPreviewFile: (relativePath: string, displayName: string) => void
+  onRetryDirectory: (relativePath: string) => void
+  onToggleDirectory: (relativePath: string) => void
+}
+
+export function MobileFileExplorerRow(props: Props) {
+  const { item, expanded, onPreviewFile, onRetryDirectory, onToggleDirectory } = props
+
+  if (item.kind === 'loading') {
+    return (
+      <View style={[styles.inlineStatusRow, { paddingLeft: spacing.lg + item.depth * 18 }]}>
+        <View style={styles.chevronSpacer} />
+        <ActivityIndicator size="small" color={colors.textSecondary} />
+        <Text style={styles.inlineStatusText}>Loading...</Text>
+      </View>
+    )
+  }
+
+  if (item.kind === 'error') {
+    return (
+      <View style={[styles.inlineStatusRow, { paddingLeft: spacing.lg + item.depth * 18 }]}>
+        <View style={styles.chevronSpacer} />
+        <Text style={styles.inlineErrorText} numberOfLines={1}>
+          {item.message || 'Unable to load folder'}
+        </Text>
+        <Pressable
+          style={({ pressed }) => [styles.inlineRetryButton, pressed && styles.rowPressed]}
+          onPress={() => {
+            triggerSelection()
+            onRetryDirectory(item.relativePath)
+          }}
+          accessibilityLabel={`Retry loading ${item.relativePath}`}
+        >
+          <Text style={styles.inlineRetryText}>Retry</Text>
+        </Pressable>
+      </View>
+    )
+  }
+
+  if (isTreeNode(item)) {
+    return (
+      <TreeRow
+        item={item}
+        expanded={expanded}
+        onPreviewFile={onPreviewFile}
+        onToggleDirectory={onToggleDirectory}
+      />
+    )
+  }
+
+  return null
+}
+
+function isTreeNode(item: FileExplorerRow): item is TreeNode {
+  return item.kind === 'directory' || item.kind === 'text' || item.kind === 'binary'
+}
+
+function TreeRow(props: {
+  item: TreeNode
+  expanded: ReadonlySet<string>
+  onPreviewFile: (relativePath: string, displayName: string) => void
+  onToggleDirectory: (relativePath: string) => void
+}) {
+  const { item, expanded, onPreviewFile, onToggleDirectory } = props
+  const isDirectory = item.kind === 'directory'
+  const isExpanded = expanded.has(item.relativePath)
+  // Images render in the mobile viewer (via files.readPreview), so a binary
+  // image is openable; only non-previewable binaries are unavailable.
+  const previewable =
+    item.kind !== 'directory' &&
+    canPreviewMobileFileRow({ kind: item.kind, relativePath: item.relativePath })
+  const isImage = item.kind === 'binary' && previewable
+  const disabled = item.kind === 'binary' && !previewable
+  const markdown = item.kind === 'text' && isMarkdownPath(item.relativePath)
+
+  return (
+    <Pressable
+      style={({ pressed }) => [
+        styles.row,
+        { paddingLeft: spacing.lg + item.depth * 18 },
+        pressed && !disabled && styles.rowPressed,
+        disabled && styles.rowDisabled
+      ]}
+      disabled={disabled}
+      onPress={() => {
+        triggerSelection()
+        if (isDirectory) {
+          onToggleDirectory(item.relativePath)
+        } else if (!disabled) {
+          onPreviewFile(item.relativePath, item.name)
+        }
+      }}
+      accessibilityLabel={
+        isDirectory
+          ? `Open folder ${item.name}`
+          : disabled
+            ? `${item.name} unavailable on mobile`
+            : `Preview file ${item.name}`
+      }
+    >
+      {isDirectory ? (
+        isExpanded ? (
+          <ChevronDown size={16} color={colors.textSecondary} />
+        ) : (
+          <ChevronRight size={16} color={colors.textSecondary} />
+        )
+      ) : (
+        <View style={styles.chevronSpacer} />
+      )}
+      {isDirectory ? (
+        <Folder size={17} color={colors.textSecondary} />
+      ) : markdown ? (
+        <FileText size={17} color={disabled ? colors.textMuted : colors.textSecondary} />
+      ) : isImage ? (
+        <ImageIcon size={17} color={colors.textSecondary} />
+      ) : (
+        <File size={17} color={disabled ? colors.textMuted : colors.textSecondary} />
+      )}
+      <View style={styles.rowTextBlock}>
+        <Text style={[styles.rowTitle, disabled && styles.rowTitleDisabled]} numberOfLines={1}>
+          {item.name}
+        </Text>
+        {disabled ? <Text style={styles.rowMeta}>Unavailable on mobile</Text> : null}
+      </View>
+    </Pressable>
+  )
+}

mobile/src/files/mobile-file-explorer-styles.ts

@@ -78,6 +78,37 @@ export const fileExplorerStyles = StyleSheet.create({
     color: colors.textMuted,
     fontSize: 11
   },
+  inlineStatusRow: {
+    minHeight: 36,
+    flexDirection: 'row',
+    alignItems: 'center',
+    gap: spacing.sm,
+    paddingRight: spacing.md
+  },
+  inlineStatusText: {
+    color: colors.textSecondary,
+    fontSize: typography.metaSize
+  },
+  inlineErrorText: {
+    flex: 1,
+    minWidth: 0,
+    color: colors.statusRed,
+    fontSize: typography.metaSize
+  },
+  inlineRetryButton: {
+    minHeight: 28,
+    alignItems: 'center',
+    justifyContent: 'center',
+    borderRadius: radii.button,
+    borderWidth: StyleSheet.hairlineWidth,
+    borderColor: colors.borderSubtle,
+    paddingHorizontal: spacing.md
+  },
+  inlineRetryText: {
+    color: colors.textPrimary,
+    fontSize: typography.metaSize,
+    fontWeight: '600'
+  },
   state: {
     flex: 1,
     alignItems: 'center',

mobile/src/files/mobile-file-preview-route.ts

@@ -111,5 +111,5 @@ export function createMobileFilePreviewHref(
 }
 
 export function displayNameFromPreviewPath(relativePath: string): string {
-  return relativePath.split(/[\\/]/).filter(Boolean).at(-1) ?? relativePath
+  return relativePath.split(/[\\/]/).findLast(Boolean) ?? relativePath
 }

mobile/src/hooks/use-mobile-dictation.ts

@@ -1,6 +1,3 @@
-/* oxlint-disable max-lines -- Why: mobile dictation keeps permission, recording,
- * chunk upload, completion, and cancellation in one hook so native audio state
- * cannot drift from the runtime RPC lifecycle. */
 import { useCallback, useEffect, useLayoutEffect, useRef, useState } from 'react'
 // Why: import from 'buffer' (the npm polyfill), not 'node:buffer' — Metro
 // can't resolve Node's builtin in a React Native bundle.

mobile/src/mobile-release/prepare-android-release-script.test.ts

@@ -0,0 +1,92 @@
+import { execFileSync, spawnSync } from 'node:child_process'
+import { mkdtempSync, readFileSync, rmSync, writeFileSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import { join } from 'node:path'
+import { fileURLToPath } from 'node:url'
+import { afterEach, describe, expect, it } from 'vitest'
+
+const scriptPath = fileURLToPath(
+  new URL('../../scripts/prepare-android-release.mjs', import.meta.url)
+)
+
+const appConfig = {
+  expo: {
+    version: '0.0.22',
+    android: {
+      versionCode: 4
+    }
+  }
+}
+
+let tempDirs: string[] = []
+
+function createAppConfig() {
+  const dir = mkdtempSync(join(tmpdir(), 'orca-android-release-'))
+  tempDirs.push(dir)
+  const configPath = join(dir, 'app.json')
+  const contents = `${JSON.stringify(appConfig, null, 2)}\n`
+  writeFileSync(configPath, contents)
+  return { configPath, contents }
+}
+
+describe('prepare Android release script', () => {
+  afterEach(() => {
+    for (const dir of tempDirs) {
+      rmSync(dir, { force: true, recursive: true })
+    }
+    tempDirs = []
+  })
+
+  it('uses committed Android release identity without mutating app config', () => {
+    const { configPath, contents } = createAppConfig()
+
+    const output = execFileSync(process.execPath, [scriptPath], {
+      encoding: 'utf8',
+      env: {
+        ...process.env,
+        MOBILE_APP_CONFIG_PATH: configPath,
+        MOBILE_ANDROID_PUBLISH_RELEASE: 'true'
+      }
+    })
+
+    expect(output).toContain('Prepared Orca Mobile Android 0.0.22 (4)')
+    expect(output).toContain('Release tag: mobile-android-v0.0.22')
+    expect(readFileSync(configPath, 'utf8')).toBe(contents)
+  })
+
+  it('rejects release-only Android versionCode bumps', () => {
+    const { configPath } = createAppConfig()
+
+    const result = spawnSync(process.execPath, [scriptPath], {
+      encoding: 'utf8',
+      env: {
+        ...process.env,
+        MOBILE_APP_CONFIG_PATH: configPath,
+        MOBILE_ANDROID_BUMP_VERSION_CODE: 'true'
+      }
+    })
+
+    expect(result.status).toBe(1)
+    expect(result.stderr).toContain(
+      'Android versionCode changes must be committed in mobile/app.json before release'
+    )
+  })
+
+  it('rejects release versions that do not match committed app config', () => {
+    const { configPath } = createAppConfig()
+
+    const result = spawnSync(process.execPath, [scriptPath], {
+      encoding: 'utf8',
+      env: {
+        ...process.env,
+        MOBILE_APP_CONFIG_PATH: configPath,
+        MOBILE_ANDROID_RELEASE_VERSION: '0.0.23'
+      }
+    })
+
+    expect(result.status).toBe(1)
+    expect(result.stderr).toContain(
+      'MOBILE_ANDROID_RELEASE_VERSION must match the committed mobile app version'
+    )
+  })
+})

mobile/src/notifications/mobile-notifications.test.ts

@@ -1,6 +1,9 @@
 import { beforeEach, describe, expect, it, vi } from 'vitest'
 import * as Notifications from 'expo-notifications'
-import { subscribeToDesktopNotifications } from './mobile-notifications'
+import {
+  setScheduledNotificationsMaxForTests,
+  subscribeToDesktopNotifications
+} from './mobile-notifications'
 import type { RpcClient } from '../transport/rpc-client'
 import { loadPushNotificationsEnabled } from '../storage/preferences'
 
@@ -268,4 +271,48 @@ describe('subscribeToDesktopNotifications', () => {
 
     expect(Notifications.dismissNotificationAsync).not.toHaveBeenCalled()
   })
+
+  // Why: notificationId is unique per completion, so the map grew unbounded when
+  // the desktop never sent a dismiss (the remote-mobile case). It is now capped.
+  it('evicts the oldest scheduled entry once the cap is exceeded', async () => {
+    setScheduledNotificationsMaxForTests(1)
+    try {
+      vi.mocked(loadPushNotificationsEnabled).mockResolvedValue(true)
+      vi.mocked(Notifications.getPermissionsAsync).mockResolvedValue({
+        status: 'granted',
+        canAskAgain: true
+      } as never)
+      vi.mocked(Notifications.scheduleNotificationAsync)
+        .mockResolvedValueOnce('scheduled-old')
+        .mockResolvedValueOnce('scheduled-new')
+      vi.mocked(Notifications.dismissNotificationAsync).mockResolvedValue(undefined)
+      let onEvent: ((data: unknown) => void) | null = null
+      const client = {
+        subscribe: vi.fn((_method, _params, callback: (data: unknown) => void) => {
+          onEvent = callback
+          return vi.fn()
+        }),
+        getState: vi.fn(() => 'connected'),
+        sendRequest: vi.fn()
+      } as unknown as RpcClient
+
+      subscribeToDesktopNotifications(client, 'host-1')
+      onEvent?.({ type: 'notification', title: 't', body: 'b', notificationId: 'agent:old' })
+      await flushAsync()
+      onEvent?.({ type: 'notification', title: 't', body: 'b', notificationId: 'agent:new' })
+      await flushAsync()
+
+      // The older entry was evicted by the cap: dismissing it is a no-op...
+      onEvent?.({ type: 'dismiss', notificationId: 'agent:old' })
+      await flushAsync()
+      expect(Notifications.dismissNotificationAsync).not.toHaveBeenCalledWith('scheduled-old')
+
+      // ...while the most-recent entry is retained and still dismissable.
+      onEvent?.({ type: 'dismiss', notificationId: 'agent:new' })
+      await flushAsync()
+      expect(Notifications.dismissNotificationAsync).toHaveBeenCalledWith('scheduled-new')
+    } finally {
+      setScheduledNotificationsMaxForTests()
+    }
+  })
 })

mobile/src/notifications/mobile-notifications.ts

@@ -31,10 +31,42 @@ type ScheduledNotificationState = {
 
 const scheduledNotificationsByHostAndNotificationId = new Map<string, ScheduledNotificationState>()
 
+// Why: notificationId embeds a per-completion timestamp (buildAgentNotificationId),
+// so every agent-task-complete inserts a new, never-reused key. Entries are only
+// removed when the desktop sends a matching dismiss — which a remote mobile user
+// (not at the desktop) frequently never gets — so the map grew for the app's whole
+// life. Bound it; a settled entry only retains a small identifier used for later
+// programmatic dismissal, unnecessary for long-past completions.
+const MAX_SCHEDULED_NOTIFICATIONS = 256
+let maxScheduledNotifications = MAX_SCHEDULED_NOTIFICATIONS
+
 function getStoredNotificationKey(hostId: string, notificationId: string): string {
   return `${encodeURIComponent(hostId)}:${encodeURIComponent(notificationId)}`
 }
 
+// Evict the oldest SETTLED entries (never one mid-schedule) until within the cap.
+// Map iteration is insertion order, so the first match is the oldest.
+function boundScheduledNotifications(): void {
+  while (scheduledNotificationsByHostAndNotificationId.size > maxScheduledNotifications) {
+    let evicted = false
+    for (const [key, state] of scheduledNotificationsByHostAndNotificationId) {
+      if (!state.pending) {
+        scheduledNotificationsByHostAndNotificationId.delete(key)
+        evicted = true
+        break
+      }
+    }
+    if (!evicted) {
+      break
+    }
+  }
+}
+
+/** Test-only: override the cap (pass no arg to restore the default). */
+export function setScheduledNotificationsMaxForTests(max?: number): void {
+  maxScheduledNotifications = max ?? MAX_SCHEDULED_NOTIFICATIONS
+}
+
 export type NotificationPermissionState = {
   granted: boolean
   status: string
@@ -155,6 +187,7 @@ async function showLocalNotification(event: NotificationEvent, hostId: string):
       return
     }
     notificationState.identifier = scheduledIdentifier
+    boundScheduledNotifications()
   } finally {
     if (notificationState.pending === pending) {
       notificationState.pending = undefined

mobile/src/session/MobileTerminalLiveInputStatus.tsx

@@ -27,9 +27,11 @@ export function MobileTerminalLiveInputStatus({
     ? 'Tap mic to stop'
     : dictation.isProcessing
       ? 'Transcribing on desktop'
-      : isAttaching
-        ? 'Uploading image to host'
-        : 'Keyboard goes directly to terminal'
+      : dictation.isStarting
+        ? 'Preparing microphone'
+        : isAttaching
+          ? 'Uploading image to host'
+          : 'Tap to show keyboard'
 
   return (
     <View style={styles.status}>

mobile/src/session/mobile-diff-comments.ts

@@ -77,7 +77,7 @@ export function normalizeMobileDiffComments(value: unknown, worktreeId: string):
     }
     const id = typeof candidate.id === 'string' ? candidate.id : ''
     const filePath = typeof candidate.filePath === 'string' ? candidate.filePath : ''
-    const lineNumber = typeof candidate.lineNumber === 'number' ? candidate.lineNumber : NaN
+    const lineNumber = typeof candidate.lineNumber === 'number' ? candidate.lineNumber : Number.NaN
     const body = typeof candidate.body === 'string' ? candidate.body.trim() : ''
     const createdAt = typeof candidate.createdAt === 'number' ? candidate.createdAt : Date.now()
     if (!id || !filePath || !Number.isFinite(lineNumber) || lineNumber < 0 || !body) {

mobile/src/session/mobile-terminal-file-tap-open.ts

@@ -169,5 +169,5 @@ function scheduleOpenedWorktreeTabActivation<T extends TerminalFileTapSessionTab
 }
 
 function displayNameFromPath(path: string): string | undefined {
-  return path.split(/[\\/]/).filter(Boolean).at(-1)
+  return path.split(/[\\/]/).findLast(Boolean)
 }

mobile/src/source-control/MobileSourceControlContent.tsx

@@ -16,6 +16,7 @@ type Props = {
 export function MobileSourceControlContent({ state }: Props) {
   const {
     insets,
+    connState,
     busyAction,
     commitMessage,
     setCommitMessage,
@@ -56,6 +57,15 @@ export function MobileSourceControlContent({ state }: Props) {
 
   return (
     <>
+      {connState !== 'connected' ? (
+        // Why: once data has loaded the screen looks alive even when the
+        // desktop link is down, so taps appear to do nothing (STA-1511).
+        // Surface the reconnect state where the user is looking.
+        <View style={styles.reconnectBanner}>
+          <ActivityIndicator size="small" color={colors.statusAmber} />
+          <Text style={styles.reconnectBannerText}>Reconnecting to desktop...</Text>
+        </View>
+      ) : null}
       <View style={styles.summaryCard}>
         <View style={styles.summaryHeader}>
           <View style={styles.branchLine}>

mobile/src/source-control/mobile-history-screen-state.test.ts

@@ -0,0 +1,47 @@
+import { describe, expect, it } from 'vitest'
+import { resolveMobileHistoryScreenView } from './mobile-history-screen-state'
+import type { MobileCommitRow } from './mobile-git-history'
+
+const row: MobileCommitRow = {
+  id: 'abc123',
+  shortId: 'abc123',
+  subject: 'commit',
+  author: 'author',
+  parentId: null,
+  relativeTime: '1m'
+}
+
+describe('resolveMobileHistoryScreenView', () => {
+  it('shows waiting (not an endless spinner) when disconnected with nothing loaded', () => {
+    // Regression (STA-1511): the load effect no-ops while disconnected, so
+    // rows stay null forever — the screen must offer a retry, not a spinner.
+    expect(resolveMobileHistoryScreenView({ connected: false, rows: null, error: null })).toEqual({
+      kind: 'waiting'
+    })
+  })
+
+  it('spins only while connected and loading', () => {
+    expect(resolveMobileHistoryScreenView({ connected: true, rows: null, error: null })).toEqual({
+      kind: 'loading'
+    })
+  })
+
+  it('keeps already-loaded rows visible across a connection drop', () => {
+    expect(resolveMobileHistoryScreenView({ connected: false, rows: [row], error: null })).toEqual({
+      kind: 'rows',
+      rows: [row]
+    })
+  })
+
+  it('prefers the error state over waiting so failures stay actionable', () => {
+    expect(resolveMobileHistoryScreenView({ connected: false, rows: null, error: 'boom' })).toEqual(
+      { kind: 'error', message: 'boom' }
+    )
+  })
+
+  it('distinguishes an empty history from a pending load', () => {
+    expect(resolveMobileHistoryScreenView({ connected: true, rows: [], error: null })).toEqual({
+      kind: 'empty'
+    })
+  })
+})

mobile/src/source-control/mobile-history-screen-state.ts

@@ -0,0 +1,30 @@
+import type { MobileCommitRow } from './mobile-git-history'
+
+export type MobileHistoryScreenView =
+  | { kind: 'error'; message: string }
+  | { kind: 'waiting' }
+  | { kind: 'loading' }
+  | { kind: 'empty' }
+  | { kind: 'rows'; rows: MobileCommitRow[] }
+
+// Why: the history load effect no-ops while the client isn't connected, so
+// without an explicit 'waiting' view the screen showed an infinite spinner
+// with no retry when opened during a reconnect window (STA-1511). Rows loaded
+// before a drop stay visible — stale history beats a blank screen.
+export function resolveMobileHistoryScreenView(input: {
+  connected: boolean
+  rows: MobileCommitRow[] | null
+  error: string | null
+}): MobileHistoryScreenView {
+  const { connected, rows, error } = input
+  if (error) {
+    return { kind: 'error', message: error }
+  }
+  if (rows !== null) {
+    return rows.length === 0 ? { kind: 'empty' } : { kind: 'rows', rows }
+  }
+  if (!connected) {
+    return { kind: 'waiting' }
+  }
+  return { kind: 'loading' }
+}

mobile/src/source-control/mobile-source-control-styles.ts

@@ -125,6 +125,24 @@ const baseStyles = StyleSheet.create({
     fontWeight: '600',
     textTransform: 'capitalize'
   },
+  reconnectBanner: {
+    flexDirection: 'row',
+    alignItems: 'center',
+    gap: spacing.sm,
+    marginHorizontal: spacing.lg,
+    marginTop: spacing.lg,
+    marginBottom: -spacing.sm,
+    paddingHorizontal: spacing.md,
+    paddingVertical: spacing.sm,
+    borderRadius: radii.button,
+    backgroundColor: colors.bgRaised,
+    borderWidth: StyleSheet.hairlineWidth,
+    borderColor: colors.statusAmber
+  },
+  reconnectBannerText: {
+    color: colors.textPrimary,
+    fontSize: typography.metaSize
+  },
   actionError: {
     marginTop: spacing.sm,
     paddingHorizontal: spacing.md,

mobile/src/terminal/TerminalWebView.tsx

@@ -1,76 +1,30 @@
 import { useRef, useCallback, forwardRef, useImperativeHandle, useEffect, useMemo } from 'react'
-import { StyleSheet, type StyleProp, type ViewStyle } from 'react-native'
+import { View, type StyleProp, type ViewStyle } from 'react-native'
 import { WebView } from 'react-native-webview'
 import type { WebViewMessageEvent } from 'react-native-webview'
-import type { RuntimeMobileTerminalTheme } from '../../../src/shared/runtime-types'
-import { colors } from '../theme/mobile-theme'
 import type { TerminalOscLinkRange } from './terminal-osc-link-ranges'
-import { XTERM_HTML } from './terminal-webview-html'
+import type {
+  MobileTerminalTheme,
+  TerminalSelectionEvents,
+  TerminalWebViewHandle
+} from './terminal-webview-contract'
+import {
+  TerminalWebViewEngineErrorOverlay,
+  useTerminalWebViewEngineErrorState
+} from './terminal-webview-engine-error-state'
+import { TERMINAL_WEBVIEW_FRAME_STYLES } from './terminal-webview-frame-styles'
+import { useTerminalWebReadyWatchdog } from './terminal-webview-ready-watchdog'
+import { XTERM_WEBVIEW_SOURCE } from './terminal-webview-html'
 import type { TerminalWebViewCommand } from './terminal-webview-messages'
 import { createTerminalWebViewPendingMessages } from './terminal-webview-pending-messages'
 
-type TerminalMouseTrackingMode = 'none' | 'x10' | 'vt200' | 'drag' | 'any'
-type TerminalOscLinks = TerminalOscLinkRange[]
-
-export type TerminalModes = {
-  bracketedPasteMode: boolean
-  altScreen: boolean
-  mouseTrackingMode: TerminalMouseTrackingMode
-  sgrMouseMode: boolean
-  sgrMousePixelsMode: boolean
-}
-
-export type TerminalKeyboardAvoidanceMetrics = {
-  cursorY: number
-  rows: number
-  altScreen: boolean
-}
-
-export type MobileTerminalTheme = RuntimeMobileTerminalTheme
-
-export type TerminalSelectionEvents = {
-  onSelectionMode?: (active: boolean) => void
-  onSelectionCopy?: (text: string) => void
-  onSelectionEvicted?: () => void
-  onModesChanged?: (modes: TerminalModes) => void
-  onKeyboardAvoidanceMetrics?: (metrics: TerminalKeyboardAvoidanceMetrics) => void
-  onHaptic?: (kind: 'selection' | 'success' | 'error' | 'edge-bump') => void
-  onTerminalInput?: (bytes: string) => void
-  onTerminalTap?: () => void
-  // Tap landed on a detected file path; RN resolves + opens it.
-  onFileTap?: (pathText: string, line: number | null, column: number | null) => void
-  // WebView-detected URL tap; RN chooses the mobile routing destination.
-  onOpenUrl?: (url: string) => void
-  // Why: pinch-to-zoom in the terminal snaps to a text-size preset and reports it
-  // here so the app persists it and keeps Settings + other panes in sync.
-  onTextScaleChange?: (scale: number) => void
-}
-
-export type TerminalWebViewHandle = {
-  write: (data: string) => void
-  init: (
-    cols: number,
-    rows: number,
-    initialData?: string,
-    preserveScroll?: boolean,
-    oscLinks?: TerminalOscLinks
-  ) => void
-  resize: (cols: number, rows: number) => void
-  // Why: reflow the local xterm buffer (scrollback included) to a new width
-  // after a server-side PTY reflow, so older wrapped lines rewrap to match the
-  // latest output. No-op on the alternate screen.
-  reflow: (cols: number, rows: number) => void
-  clear: () => void
-  measureFitDimensions: (containerHeight?: number) => Promise<{ cols: number; rows: number } | null>
-  resetZoom: () => void
-  cancelSelect: () => void
-  doSelectAll: () => void
-  // Why: lets callers await the WebView-side `init` rAF chain (term.open
-  // → renderService population → first paint) so a follow-up measure
-  // doesn't race ahead and find term=null or cellWidth=0. Resolves on
-  // the next 'ready' notify after the most recent init.
-  awaitReady: () => Promise<void>
-}
+export type {
+  MobileTerminalTheme,
+  TerminalKeyboardAvoidanceMetrics,
+  TerminalModes,
+  TerminalSelectionEvents,
+  TerminalWebViewHandle
+} from './terminal-webview-contract'
 
 type Props = {
   style?: StyleProp<ViewStyle>
@@ -79,18 +33,16 @@ type Props = {
   // scale; raw xterm fontSize can't drive apparent size because the fit cancels it.
   textScale?: number
   onWebReady?: () => void
+  onEngineError?: (message: string) => void
 } & TerminalSelectionEvents
 
-// Why: WebView treats source identity as page identity on some platforms; keep
-// parent/session re-renders from reloading xterm and forcing fresh snapshots.
-const XTERM_WEBVIEW_SOURCE = { html: XTERM_HTML }
-
 export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function TerminalWebView(
   {
     style,
     terminalTheme,
     textScale = 1,
     onWebReady,
+    onEngineError,
     onSelectionMode,
     onSelectionCopy,
     onSelectionEvicted,
@@ -119,6 +71,12 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
   // race ahead of term.open() / renderService population.
   const readyPromiseRef = useRef<Promise<void> | null>(null)
   const readyResolveRef = useRef<(() => void) | null>(null)
+  const { clearEngineError, engineError, reportEngineError, reportNativeEngineError } =
+    useTerminalWebViewEngineErrorState(onEngineError)
+  const { armWebReadyWatchdog, clearWebReadyWatchdog } = useTerminalWebReadyWatchdog(
+    isWebReadyRef,
+    reportEngineError
+  )
 
   const sendToWebView = useCallback((msg: TerminalWebViewCommand) => {
     messageIdRef.current += 1
@@ -151,6 +109,8 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
 
       if (msg.type === 'web-ready') {
         isWebReadyRef.current = true
+        clearWebReadyWatchdog()
+        clearEngineError()
         onWebReady?.()
         flushPendingMessages()
       } else if (msg.type === 'ready') {
@@ -175,6 +135,9 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
         const tag = typeof msg.tag === 'string' ? msg.tag : '[fit]'
         // eslint-disable-next-line no-console
         console.log(tag, msg.payload)
+      } else if (msg.type === 'error') {
+        const message = typeof msg.message === 'string' ? msg.message : 'Unknown terminal error'
+        reportEngineError(message, msg.fatal !== false)
       } else if (msg.type === 'set-select-mode') {
         onSelectionMode?.(!!msg.enabled)
       } else if (msg.type === 'selection') {
@@ -246,6 +209,9 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
     },
     [
       flushPendingMessages,
+      clearEngineError,
+      clearWebReadyWatchdog,
+      reportEngineError,
       onWebReady,
       onSelectionMode,
       onSelectionCopy,
@@ -263,10 +229,16 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
 
   const handleLoadStart = useCallback(() => {
     isWebReadyRef.current = false
+    armWebReadyWatchdog()
     // Why: messages queued for a previous WebView generation are stale after a reload;
     // dropping them avoids replaying terminal chunks before the next init snapshot.
     pendingMessages.clear()
-  }, [pendingMessages])
+  }, [armWebReadyWatchdog, pendingMessages])
+
+  const handleReload = useCallback(() => {
+    clearEngineError()
+    webViewRef.current?.reload()
+  }, [clearEngineError])
 
   useEffect(() => {
     postMessage({ type: 'set-theme', terminalTheme })
@@ -289,7 +261,7 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
         rows: number,
         initialData?: string,
         preserveScroll?: boolean,
-        oscLinks?: TerminalOscLinks
+        oscLinks?: TerminalOscLinkRange[]
       ) {
         // Why: arm a fresh ready promise BEFORE posting init. The WebView
         // resolves it via the 'ready' notify at the end of its rAF chain.
@@ -395,29 +367,35 @@ export const TerminalWebView = forwardRef<TerminalWebViewHandle, Props>(function
   )
 
   return (
-    <WebView
-      ref={webViewRef}
-      source={XTERM_WEBVIEW_SOURCE}
-      style={[styles.webview, style]}
-      originWhitelist={['*']}
-      javaScriptEnabled
-      scrollEnabled={false}
-      // Why: Android parent gesture containers can intercept vertical drags
-      // before the injected xterm scroll router sees them.
-      nestedScrollEnabled
-      scalesPageToFit={false}
-      // Why: Android WebView defaults textZoom to the system font scale, inflating
-      // xterm's DOM glyphs past its canvas-measured cell grid (#4579). iOS ignores it.
-      textZoom={100}
-      onLoadStart={handleLoadStart}
-      onMessage={handleMessage}
-    />
+    <View style={[TERMINAL_WEBVIEW_FRAME_STYLES.container, style]}>
+      <WebView
+        ref={webViewRef}
+        source={XTERM_WEBVIEW_SOURCE}
+        style={TERMINAL_WEBVIEW_FRAME_STYLES.webview}
+        originWhitelist={['*']}
+        javaScriptEnabled
+        scrollEnabled={false}
+        // Why: Android parent gesture containers can intercept vertical drags
+        // before the injected xterm scroll router sees them.
+        nestedScrollEnabled
+        scalesPageToFit={false}
+        // Why: Android WebView defaults textZoom to the system font scale, inflating
+        // xterm's DOM glyphs past its canvas-measured cell grid (#4579). iOS ignores it.
+        textZoom={100}
+        onLoadStart={handleLoadStart}
+        onMessage={handleMessage}
+        onError={(event) => reportNativeEngineError('Terminal WebView load failed', event)}
+        onHttpError={(event) => reportNativeEngineError('Terminal WebView HTTP error', event)}
+        onRenderProcessGone={(event) =>
+          reportNativeEngineError('Terminal WebView render process ended', event)
+        }
+        onContentProcessDidTerminate={(event) =>
+          reportNativeEngineError('Terminal WebView content process ended', event)
+        }
+      />
+      {engineError ? (
+        <TerminalWebViewEngineErrorOverlay message={engineError} onReload={handleReload} />
+      ) : null}
+    </View>
   )
 })
-
-const styles = StyleSheet.create({
-  webview: {
-    flex: 1,
-    backgroundColor: colors.terminalBg
-  }
-})

mobile/src/terminal/terminal-ios-dictation-write-back.test.ts

@@ -0,0 +1,25 @@
+import { readFileSync } from 'node:fs'
+import { describe, expect, it } from 'vitest'
+
+const sessionRouteSource = readFileSync(
+  new URL('../../app/h/[hostId]/session/[worktreeId].tsx', import.meta.url),
+  'utf8'
+)
+
+// Why: iOS terminates an active keyboard-dictation (and IME) session whenever
+// JS writes a value into the focused field that differs from the native text
+// (RN applies it via setTextAndSelection / _setAttributedString). Terminal
+// inputs therefore must echo the raw field text in their controlled value and
+// apply dash normalization only on the send/mirror path. See stablyai/orca#7925.
+describe('terminal iOS dictation write-back', () => {
+  it('does not write normalized text back into the buffered command input value', () => {
+    expect(sessionRouteSource).toContain('onChangeText={setInput}')
+    expect(sessionRouteSource).not.toContain(
+      'setInput((previousText) => normalizeTerminalTextInput'
+    )
+  })
+
+  it('still normalizes the buffered command text at send time', () => {
+    expect(sessionRouteSource).toContain('normalizeTerminalTextInput(input)')
+  })
+})

mobile/src/terminal/terminal-keyboard-dismiss.test.ts

@@ -0,0 +1,82 @@
+import { describe, expect, it, vi } from 'vitest'
+
+import { dismissTerminalKeyboard } from './terminal-keyboard-dismiss'
+
+describe('dismissTerminalKeyboard', () => {
+  it('clears pending live input focus before blur and dismiss', () => {
+    const calls: string[] = []
+    const clearPendingLiveInputFocus = vi.fn(() => calls.push('clear'))
+    const liveInput = { blur: vi.fn(() => calls.push('live-blur')) }
+    const commandInput = { blur: vi.fn(() => calls.push('command-blur')) }
+    const dismissKeyboard = vi.fn(() => calls.push('dismiss'))
+
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus,
+      commandInput,
+      dismissKeyboard,
+      liveInput
+    })
+
+    expect(calls).toEqual(['clear', 'live-blur', 'command-blur', 'dismiss'])
+  })
+
+  it('blurs both live and buffered command inputs', () => {
+    const liveInput = { blur: vi.fn() }
+    const commandInput = { blur: vi.fn() }
+
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus: vi.fn(),
+      commandInput,
+      dismissKeyboard: vi.fn(),
+      liveInput
+    })
+
+    expect(liveInput.blur).toHaveBeenCalledTimes(1)
+    expect(commandInput.blur).toHaveBeenCalledTimes(1)
+  })
+
+  it('dismisses the keyboard without a live input handle', () => {
+    const commandInput = { blur: vi.fn() }
+    const dismissKeyboard = vi.fn()
+
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus: vi.fn(),
+      commandInput,
+      dismissKeyboard,
+      liveInput: null
+    })
+
+    expect(commandInput.blur).toHaveBeenCalledTimes(1)
+    expect(dismissKeyboard).toHaveBeenCalledTimes(1)
+  })
+
+  it('dismisses the keyboard without a buffered command input handle', () => {
+    const liveInput = { blur: vi.fn() }
+    const dismissKeyboard = vi.fn()
+
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus: vi.fn(),
+      commandInput: undefined,
+      dismissKeyboard,
+      liveInput
+    })
+
+    expect(liveInput.blur).toHaveBeenCalledTimes(1)
+    expect(dismissKeyboard).toHaveBeenCalledTimes(1)
+  })
+
+  it('still clears focus and dismisses when both input handles are missing', () => {
+    const calls: string[] = []
+    const clearPendingLiveInputFocus = vi.fn(() => calls.push('clear'))
+    const dismissKeyboard = vi.fn(() => calls.push('dismiss'))
+
+    dismissTerminalKeyboard({
+      clearPendingLiveInputFocus,
+      commandInput: undefined,
+      dismissKeyboard,
+      liveInput: null
+    })
+
+    expect(calls).toEqual(['clear', 'dismiss'])
+  })
+})

mobile/src/terminal/terminal-keyboard-dismiss.ts

@@ -0,0 +1,17 @@
+export type TerminalKeyboardDismissHandle = { blur: () => void } | null | undefined
+
+export type DismissTerminalKeyboardOptions = {
+  clearPendingLiveInputFocus: () => void
+  commandInput: TerminalKeyboardDismissHandle
+  dismissKeyboard: () => void
+  liveInput: TerminalKeyboardDismissHandle
+}
+
+export function dismissTerminalKeyboard(options: DismissTerminalKeyboardOptions): void {
+  // Why: clear the queued live-input focus before blurring/dismissing so a
+  // pending deferred focus cannot re-open the iOS keyboard right after Hide.
+  options.clearPendingLiveInputFocus()
+  options.liveInput?.blur()
+  options.commandInput?.blur()
+  options.dismissKeyboard()
+}

mobile/src/terminal/terminal-live-hangul-mirror.test.ts

@@ -0,0 +1,178 @@
+import { describe, expect, it } from 'vitest'
+import {
+  buildTerminalLiveMirrorPayload,
+  computeTerminalLiveMirrorStep,
+  isTerminalLiveHangulCodePoint,
+  type TerminalLiveMirrorStep
+} from './terminal-live-hangul-mirror'
+
+type MirrorRun = {
+  readonly payloads: readonly string[]
+  readonly sentText: string
+  readonly heldText: string
+}
+
+function runMirrorSequence(
+  fieldStates: readonly string[],
+  options: { readonly commitAtEnd: boolean } = { commitAtEnd: false }
+): MirrorRun {
+  const payloads: string[] = []
+  let sentText = ''
+  let heldText = ''
+  for (const fieldText of fieldStates) {
+    const step = computeTerminalLiveMirrorStep(sentText, fieldText, { commitHeld: false })
+    const payload = buildTerminalLiveMirrorPayload(step)
+    if (payload.length > 0) {
+      payloads.push(payload)
+    }
+    sentText = step.nextSentText
+    heldText = step.heldText
+  }
+  if (options.commitAtEnd) {
+    const lastField = sentText + heldText
+    const step = computeTerminalLiveMirrorStep(sentText, lastField, { commitHeld: true })
+    const payload = buildTerminalLiveMirrorPayload(step)
+    if (payload.length > 0) {
+      payloads.push(payload)
+    }
+    sentText = step.nextSentText
+    heldText = step.heldText
+  }
+  return { payloads, sentText, heldText }
+}
+
+describe('terminal live hangul mirror', () => {
+  it('Given single-syllable composition When steps run Then leaks no jamo and commits only the final syllable', () => {
+    // Given / When
+    const run = runMirrorSequence(['ㅎ', '하', '한'], { commitAtEnd: true })
+
+    // Then
+    expect(run.payloads).toEqual(['한'])
+    expect(run.sentText).toBe('한')
+    expect(run.heldText).toBe('')
+  })
+
+  it('Given multi-syllable composition When a new syllable starts Then streams the stable prefix without erases', () => {
+    // Given / When
+    const run = runMirrorSequence(['ㅎ', '하', '한', '한ㄱ', '한그', '한글'], { commitAtEnd: true })
+
+    // Then
+    expect(run.payloads).toEqual(['한', '글'])
+    expect(run.sentText).toBe('한글')
+  })
+
+  it('Given dubeolsik resplit 간→가나 When steps run Then never sends the intermediate syllable', () => {
+    // Given / When
+    const run = runMirrorSequence(['ㄱ', '가', '간', '가나'], { commitAtEnd: true })
+
+    // Then
+    expect(run.payloads).toEqual(['가', '나'])
+    expect(run.sentText).toBe('가나')
+  })
+
+  it('Given a timer-committed syllable When composition continues Then erases and recommits via DEL correction', () => {
+    // Given: '하' was committed by the settle timer
+    const commit = computeTerminalLiveMirrorStep('', '하', { commitHeld: true })
+    expect(buildTerminalLiveMirrorPayload(commit)).toBe('하')
+    expect(commit.nextSentText).toBe('하')
+
+    // When: user keeps composing '하' → '한'
+    const correction = computeTerminalLiveMirrorStep(commit.nextSentText, '한', {
+      commitHeld: false
+    })
+
+    // Then: one DEL erases the stale syllable; the new one is held again
+    expect(buildTerminalLiveMirrorPayload(correction)).toBe('\x7f')
+    expect(correction.nextSentText).toBe('')
+    expect(correction.heldText).toBe('한')
+
+    const recommit = computeTerminalLiveMirrorStep('', '한', { commitHeld: true })
+    expect(buildTerminalLiveMirrorPayload(recommit)).toBe('한')
+  })
+
+  it('Given pure ASCII typing When steps run Then mirrors immediately with no held text', () => {
+    // Given / When
+    const run = runMirrorSequence(['a', 'ab', 'abc'])
+
+    // Then
+    expect(run.payloads).toEqual(['a', 'b', 'c'])
+    expect(run.heldText).toBe('')
+  })
+
+  it('Given a trailing space after Hangul When the step runs Then the space commits the held syllable', () => {
+    // Given: '한글' typed, '한' streamed, '글' held
+    const beforeSpace = runMirrorSequence(['ㅎ', '하', '한', '한ㄱ', '한그', '한글'])
+    expect(beforeSpace.sentText).toBe('한')
+    expect(beforeSpace.heldText).toBe('글')
+
+    // When
+    const step = computeTerminalLiveMirrorStep(beforeSpace.sentText, '한글 ', {
+      commitHeld: false
+    })
+
+    // Then
+    expect(buildTerminalLiveMirrorPayload(step)).toBe('글 ')
+    expect(step.heldText).toBe('')
+    expect(step.nextSentText).toBe('한글 ')
+  })
+
+  it('Given a trailing ASCII letter after Hangul When the step runs Then Hangul is committed with the letter', () => {
+    // Given
+    const held = computeTerminalLiveMirrorStep('', '한', { commitHeld: false })
+    expect(held.heldText).toBe('한')
+
+    // When
+    const step = computeTerminalLiveMirrorStep(held.nextSentText, '한a', { commitHeld: false })
+
+    // Then
+    expect(buildTerminalLiveMirrorPayload(step)).toBe('한a')
+    expect(step.heldText).toBe('')
+  })
+
+  it('Given sent text When the user deletes everything Then erases with one DEL per code point', () => {
+    // Given / When
+    const step = computeTerminalLiveMirrorStep('한글a', '', { commitHeld: false })
+
+    // Then
+    expect(step).toEqual<TerminalLiveMirrorStep>({
+      eraseCount: 3,
+      appendText: '',
+      nextSentText: '',
+      heldText: ''
+    })
+    expect(buildTerminalLiveMirrorPayload(step)).toBe('\x7f\x7f\x7f')
+  })
+
+  it('Given empty field and empty sent text When committing Then produces a zero step', () => {
+    // Given / When
+    const step = computeTerminalLiveMirrorStep('', '', { commitHeld: true })
+
+    // Then
+    expect(buildTerminalLiveMirrorPayload(step)).toBe('')
+    expect(step).toEqual<TerminalLiveMirrorStep>({
+      eraseCount: 0,
+      appendText: '',
+      nextSentText: '',
+      heldText: ''
+    })
+  })
+
+  it('Given non-Hangul IME text When the step runs Then it mirrors immediately without holding', () => {
+    // Given / When
+    const chinese = computeTerminalLiveMirrorStep('', '你好', { commitHeld: false })
+    const vietnamese = computeTerminalLiveMirrorStep('', 'tiếng', { commitHeld: false })
+
+    // Then
+    expect(buildTerminalLiveMirrorPayload(chinese)).toBe('你好')
+    expect(chinese.heldText).toBe('')
+    expect(buildTerminalLiveMirrorPayload(vietnamese)).toBe('tiếng')
+    expect(vietnamese.heldText).toBe('')
+  })
+
+  it('Given Hangul code point ranges When checked Then jamo and syllables match and ASCII does not', () => {
+    expect(isTerminalLiveHangulCodePoint('ㅎ'.codePointAt(0) ?? 0)).toBe(true)
+    expect(isTerminalLiveHangulCodePoint('한'.codePointAt(0) ?? 0)).toBe(true)
+    expect(isTerminalLiveHangulCodePoint('a'.codePointAt(0) ?? 0)).toBe(false)
+    expect(isTerminalLiveHangulCodePoint('あ'.codePointAt(0) ?? 0)).toBe(false)
+  })
+})

mobile/src/terminal/terminal-live-hangul-mirror.ts

@@ -0,0 +1,61 @@
+// Why: paused composition should still reach the PTY quickly; corrections make
+// a premature commit safe, so this can be short without leaking jamo forever.
+export const TERMINAL_LIVE_HELD_SYLLABLE_COMMIT_DELAY_MS = 300
+
+const TERMINAL_DEL_BYTE = '\x7f'
+
+export function isTerminalLiveHangulCodePoint(codePoint: number): boolean {
+  return (
+    (codePoint >= 0x1100 && codePoint <= 0x11ff) ||
+    (codePoint >= 0x3130 && codePoint <= 0x318f) ||
+    (codePoint >= 0xa960 && codePoint <= 0xa97f) ||
+    (codePoint >= 0xac00 && codePoint <= 0xd7af)
+  )
+}
+
+export type TerminalLiveMirrorStep = {
+  readonly eraseCount: number
+  readonly appendText: string
+  readonly nextSentText: string
+  readonly heldText: string
+}
+
+// Why: React Native exposes no composition events, but Hangul composition only
+// mutates the trailing syllable. Holding just that code point keeps the PTY
+// echo live while preedit jamo never leak; DEL corrections repair any commit
+// that later turns out to be premature.
+export function computeTerminalLiveMirrorStep(
+  sentText: string,
+  fieldText: string,
+  options: { readonly commitHeld: boolean }
+): TerminalLiveMirrorStep {
+  const fieldCodePoints = Array.from(fieldText)
+  const lastCodePoint = fieldCodePoints.at(-1)
+  const holdLast =
+    !options.commitHeld &&
+    lastCodePoint !== undefined &&
+    isTerminalLiveHangulCodePoint(lastCodePoint.codePointAt(0) ?? 0)
+  const heldText = holdLast && lastCodePoint !== undefined ? lastCodePoint : ''
+  const targetCodePoints = holdLast ? fieldCodePoints.slice(0, -1) : fieldCodePoints
+  const sentCodePoints = Array.from(sentText)
+
+  let commonPrefixLength = 0
+  while (
+    commonPrefixLength < sentCodePoints.length &&
+    commonPrefixLength < targetCodePoints.length &&
+    sentCodePoints[commonPrefixLength] === targetCodePoints[commonPrefixLength]
+  ) {
+    commonPrefixLength += 1
+  }
+
+  return {
+    eraseCount: sentCodePoints.length - commonPrefixLength,
+    appendText: targetCodePoints.slice(commonPrefixLength).join(''),
+    nextSentText: targetCodePoints.join(''),
+    heldText
+  }
+}
+
+export function buildTerminalLiveMirrorPayload(step: TerminalLiveMirrorStep): string {
+  return TERMINAL_DEL_BYTE.repeat(step.eraseCount) + step.appendText
+}

mobile/src/terminal/terminal-live-input-affordance.test.ts

@@ -0,0 +1,49 @@
+import { readFileSync } from 'node:fs'
+import { describe, expect, it } from 'vitest'
+
+const sessionRouteSource = readFileSync(
+  new URL('../../app/h/[hostId]/session/[worktreeId].tsx', import.meta.url),
+  'utf8'
+)
+const liveInputStatusSource = readFileSync(
+  new URL('../session/MobileTerminalLiveInputStatus.tsx', import.meta.url),
+  'utf8'
+)
+const commandInputStylesSource = readFileSync(
+  new URL('../../app/h/[hostId]/session/mobile-session-command-input-styles.ts', import.meta.url),
+  'utf8'
+)
+
+function liveInputBarBlock(): string {
+  const start = sessionRouteSource.indexOf('{liveInputEnabled ? (')
+  expect(start).toBeGreaterThanOrEqual(0)
+  const end = sessionRouteSource.indexOf(') : (', start)
+  expect(end).toBeGreaterThan(start)
+  return sessionRouteSource.slice(start, end)
+}
+
+describe('terminal live input affordance', () => {
+  it('keeps the live status row wired as the keyboard focus control', () => {
+    const block = liveInputBarBlock()
+
+    expect(block).toContain('onPress={focusLiveInput}')
+    expect(block).toContain('accessibilityRole="button"')
+    expect(block).toContain('accessibilityLabel="Show keyboard for live terminal input"')
+    expect(block).toContain(
+      'accessibilityHint="Typed text is sent directly to the active terminal"'
+    )
+    expect(block).toContain('pressed && styles.liveInputFocusTargetPressed')
+    expect(block).toContain('!canSend && styles.liveInputFocusTargetDisabled')
+    expect(block).toContain('showSoftInputOnFocus')
+    expect(sessionRouteSource).toContain('focusTerminalLiveInputTarget(liveInputRef.current')
+    expect(sessionRouteSource).toContain('keyboardHeight')
+    expect(sessionRouteSource).toContain('scheduleTerminalLiveInputFocus(liveInputFocusTimerRef')
+  })
+
+  it('makes the live keyboard target visible instead of status-only chrome', () => {
+    expect(liveInputStatusSource).toContain("'Tap to show keyboard'")
+    expect(commandInputStylesSource).toContain('backgroundColor: colors.bgRaised')
+    expect(commandInputStylesSource).toContain('borderWidth: 1')
+    expect(commandInputStylesSource).toContain('liveInputFocusTargetPressed')
+  })
+})

mobile/src/terminal/terminal-live-input.test.ts

@@ -5,10 +5,12 @@ import {
   clearTerminalLiveInputFocusTimer,
   defaultTerminalLiveInputHandles,
   filterTerminalLiveInputDefaultCandidates,
+  focusTerminalLiveInputTarget,
   getTerminalLiveSpecialKeyBytes,
   isTerminalLiveInputWithinByteLimit,
   pruneTerminalLiveInputHandles,
   scheduleTerminalLiveInputFocus,
+  type TerminalLiveInputFocusTarget,
   type TerminalLiveInputFocusTimerRef
 } from './terminal-live-input'
 
@@ -16,6 +18,14 @@ function createTimerRef(): TerminalLiveInputFocusTimerRef {
   return { current: null }
 }
 
+function createFocusTarget(isFocused: () => boolean): TerminalLiveInputFocusTarget {
+  return {
+    blur: vi.fn(),
+    focus: vi.fn(),
+    isFocused
+  }
+}
+
 describe('terminal live input', () => {
   afterEach(() => {
     vi.useRealTimers()
@@ -181,4 +191,37 @@ describe('terminal live input', () => {
     expect(focus).not.toHaveBeenCalled()
     expect(timerRef.current).toBeNull()
   })
+
+  it('refocuses an already-focused capture input when the keyboard is closed', () => {
+    const input = createFocusTarget(() => true)
+    const refocus = vi.fn()
+
+    focusTerminalLiveInputTarget(input, { keyboardHeight: 0, refocus })
+
+    expect(input.blur).toHaveBeenCalledTimes(1)
+    expect(input.focus).not.toHaveBeenCalled()
+    expect(refocus).toHaveBeenCalledTimes(1)
+  })
+
+  it('focuses the capture input directly when the keyboard is open', () => {
+    const input = createFocusTarget(() => true)
+    const refocus = vi.fn()
+
+    focusTerminalLiveInputTarget(input, { keyboardHeight: 240, refocus })
+
+    expect(input.blur).not.toHaveBeenCalled()
+    expect(input.focus).toHaveBeenCalledTimes(1)
+    expect(refocus).not.toHaveBeenCalled()
+  })
+
+  it('focuses the capture input directly when it is not already focused', () => {
+    const input = createFocusTarget(() => false)
+    const refocus = vi.fn()
+
+    focusTerminalLiveInputTarget(input, { keyboardHeight: 0, refocus })
+
+    expect(input.blur).not.toHaveBeenCalled()
+    expect(input.focus).toHaveBeenCalledTimes(1)
+    expect(refocus).not.toHaveBeenCalled()
+  })
 })

mobile/src/terminal/terminal-live-input.ts

@@ -66,6 +66,17 @@ export type TerminalLiveInputFocusTimerRef = {
   current: ReturnType<typeof setTimeout> | null
 }
 
+export type TerminalLiveInputFocusTarget = {
+  readonly focus: () => void
+  readonly blur: () => void
+  readonly isFocused?: () => boolean
+}
+
+type FocusTerminalLiveInputTargetOptions = {
+  readonly keyboardHeight: number
+  readonly refocus: () => void
+}
+
 export type TerminalLiveInputDefaultResult = {
   enabledHandles: ReadonlySet<string>
   defaultedHandles: ReadonlySet<string>
@@ -217,4 +228,23 @@ export function scheduleTerminalLiveInputFocus(
   }, delayMs)
 }
 
+export function focusTerminalLiveInputTarget(
+  input: TerminalLiveInputFocusTarget | null,
+  { keyboardHeight, refocus }: FocusTerminalLiveInputTargetOptions
+): void {
+  if (!input) {
+    return
+  }
+
+  if (keyboardHeight <= 0 && input.isFocused?.()) {
+    // Why: Android can keep a hidden TextInput focused after the IME is dismissed;
+    // focus() is then a no-op, so force a new focus session to reopen the keyboard.
+    input.blur()
+    refocus()
+    return
+  }
+
+  input.focus()
+}
+
 export { TERMINAL_LIVE_INPUT_MAX_BYTES }

mobile/src/terminal/terminal-live-pending-flush-state.test.ts

@@ -1,7 +1,7 @@
 import { describe, expect, it } from 'vitest'
 import { sendTerminalLiveControlAfterPendingFlush } from './terminal-live-control-send-order'
 import {
-  queueTerminalLivePendingFlush,
+  queueTerminalLiveMirrorSend,
   waitForTerminalLivePendingFlush,
   type TerminalLivePendingFlushState
 } from './terminal-live-pending-flush-state'
@@ -64,83 +64,54 @@ describe('terminal live pending flush state', () => {
     await expect(controlSend).resolves.toBe(false)
     expect(events).toEqual([])
   })
+})
 
-  it('Given a queued pending flush When it resolves or rejects Then clears the barrier', async () => {
+describe('terminal live mirror send queue', () => {
+  it('Given a failed previous send When a mirror send queues Then it still runs in order', async () => {
     // Given
-    const resolvedState: TerminalLivePendingFlushState = { current: null }
-    const rejectedState: TerminalLivePendingFlushState = { current: null }
+    const state: TerminalLivePendingFlushState = { current: null }
+    const order: string[] = []
+    const first = queueTerminalLiveMirrorSend(state, async () => {
+      order.push('first')
+      return false
+    })
 
     // When
-    await expect(queueTerminalLivePendingFlush(resolvedState, async () => true)).resolves.toBe(true)
-    await expect(
-      queueTerminalLivePendingFlush(rejectedState, async () => {
-        throw new Error('send failed')
-      })
-    ).resolves.toBe(false)
+    const second = queueTerminalLiveMirrorSend(state, async () => {
+      order.push('second')
+      return true
+    })
+
+    // Then
+    await expect(first).resolves.toBe(false)
+    await expect(second).resolves.toBe(true)
+    expect(order).toEqual(['first', 'second'])
+  })
+
+  it('Given a throwing send When a mirror send queues Then the promise resolves false and the chain continues', async () => {
+    // Given
+    const state: TerminalLivePendingFlushState = { current: null }
+    const first = queueTerminalLiveMirrorSend(state, async () => {
+      throw new Error('boom')
+    })
+
+    // When
+    const second = queueTerminalLiveMirrorSend(state, async () => true)
+
+    // Then
+    await expect(first).resolves.toBe(false)
+    await expect(second).resolves.toBe(true)
+  })
+
+  it('Given a settled mirror send When it was the newest Then the state resets to null', async () => {
+    // Given
+    const state: TerminalLivePendingFlushState = { current: null }
+
+    // When
+    await queueTerminalLiveMirrorSend(state, async () => true)
     await Promise.resolve()
 
     // Then
-    expect(resolvedState.current).toBeNull()
-    expect(rejectedState.current).toBeNull()
-  })
-
-  it('Given a current pending snapshot while another flush is in flight When queued Then sends current text after prior success', async () => {
-    // Given
-    const events: string[] = []
-    let resolveFirstFlush: (value: boolean) => void = () => {}
-    const firstFlush = new Promise<boolean>((resolve) => {
-      resolveFirstFlush = resolve
-    })
-    const state: TerminalLivePendingFlushState = { current: firstFlush }
-
-    // When
-    const secondFlush = queueTerminalLivePendingFlush(state, async () => {
-      events.push('second-flush')
-      return true
-    })
-    const controlSend = sendTerminalLiveControlAfterPendingFlush(
-      () => waitForTerminalLivePendingFlush(state),
-      async () => {
-        events.push('control')
-        return true
-      }
-    )
-    await Promise.resolve()
-
-    // Then
-    expect(events).toEqual([])
-    resolveFirstFlush(true)
-    await expect(secondFlush).resolves.toBe(true)
-    await expect(controlSend).resolves.toBe(true)
-    expect(events).toEqual(['second-flush', 'control'])
-  })
-
-  it('Given a prior pending flush fails When another snapshot is queued Then skips the current text and control', async () => {
-    // Given
-    const events: string[] = []
-    let resolveFirstFlush: (value: boolean) => void = () => {}
-    const firstFlush = new Promise<boolean>((resolve) => {
-      resolveFirstFlush = resolve
-    })
-    const state: TerminalLivePendingFlushState = { current: firstFlush }
-
-    // When
-    const secondFlush = queueTerminalLivePendingFlush(state, async () => {
-      events.push('second-flush')
-      return true
-    })
-    const controlSend = sendTerminalLiveControlAfterPendingFlush(
-      () => waitForTerminalLivePendingFlush(state),
-      async () => {
-        events.push('control')
-        return true
-      }
-    )
-    resolveFirstFlush(false)
-
-    // Then
-    await expect(secondFlush).resolves.toBe(false)
-    await expect(controlSend).resolves.toBe(false)
-    expect(events).toEqual([])
+    expect(state.current).toBeNull()
   })
 })