chore(main): release 1.766.1 (#10263 )

* chore(main): release 1.766.1 * Apply automatic changes --------- Co-authored-by: rubenfiszel <275584+rubenfiszel@users.noreply.github.com>
fix(jobs): enforce self_approval_disabled on the UI resume path (#10262 )
2026-09-15 08:02:27 +00:00 · 2026-07-22 17:00:08 +02:00 · 2026-07-22 16:52:46 +02:00 · 2026-07-22 16:35:56 +02:00 · 2026-07-22 16:21:13 +02:00 · 2026-07-22 14:35:22 +02:00
1543 changed files with 145535 additions and 19151 deletions
@@ -31,6 +31,21 @@ The `up.sql` usually defines:
  - trigger-specific fields
 - Indexes on foreign keys + any frequently-filtered columns
 - Foreign key to `workspace`
+- The RLS policies (`see_own`, `see_member`, `see_folder_extra_perms_user_*`, `see_extra_perms_user_*`, `see_extra_perms_groups_*`), copied from an existing trigger table
+
+**RLS: wrap every session GUC read in a scalar sub-select.** Write the session
+reads as `(select current_setting('session.user'))`,
+`= any((select regexp_split_to_array(current_setting('session.groups'), ','))::text[])`,
+`?| (select regexp_split_to_array(current_setting('session.pgroups'), ','))::text[]`,
+`? (select concat('u/', current_setting('session.user')))`, etc. — not the bare
+`current_setting(...)`. The GUCs are set with `SET LOCAL`, so the sub-select
+hoists them to a one-time InitPlan instead of re-evaluating per scanned row.
+Put the `::text[]` cast **outside** the sub-select for the array cases: in an
+`= any (...)` context, casting inside — `= any((select ...::text[]))` — makes
+Postgres parse the operand as a row-returning subquery and fails at CREATE with
+`operator does not exist: text = text[]`. The outside cast keeps it in
+array-operand form. See migration `20260714230440_wrap_session_gucs_in_rls_policies`
+for the canonical wrapped forms.

 Down migration drops the table and any enum types.

@@ -0,0 +1,40 @@
+---
+name: ai-chat
+description: Guidance for improving the Windmill AI chat (copilot), especially global mode — tools, prompts, and context-window discipline. Use when editing chat tools, system prompts, or tool-result shapes under frontend/src/lib/components/copilot/chat, or when changing how the chat manages its context window.
+---
+
+## Always benchmark before and after
+
+No context or behavior change ships without an `ai_evals` A/B on the affected mode.
+Add or adjust cases for exactly what you changed — see the `ai-evals` skill for
+authoring and the full run reference.
+
+Run the affected mode **before** your change and **after**, same model(s), same cases.
+
+## Measure the window first, and cumulative second
+
+Optimize **`finalContextTokens`** (window occupancy — what drives overflow and
+compaction), then cumulative prompt tokens.
+
+## Context discipline
+
+The dominant fixed cost is per-iteration overhead: the system prompt **plus every
+tool schema** is re-sent on every loop iteration. So:
+
+- **Every tool and every parameter is a permanent tax.** Justify each one and measure
+  it; an extra "locate" round-trip can cost more than the reads it saves. Strip dead
+  params rather than leaving them in the schema.
+- **Tool results return the minimum.** Never echo content the model already has. The
+  canonical mistake: a write tool that returns the whole edited artifact right after
+  the model authored it — return `{ success, message }` instead. When you touch a
+  *shared* write helper (e.g. `finishAppDraftWrite` in `global/core.ts`), re-check
+  this invariant for **all** the write tools routing through it — the echo has
+  regressed before via a shared refactor.
+
+## Prompts and tool descriptions are part of the surface
+
+The system prompt and tool descriptions steer behavior as much as the tools
+themselves, and are benchmarkable the same way. A description that advertises
+truncation makes the model self-limit; the path-conventions block changes where
+drafts land. Treat prompt/description edits as real changes and A/B them — a
+pure-prompt change is a legitimate, measurable improvement.
@@ -0,0 +1,87 @@
+---
+name: ai-evals
+description: Author and run black-box benchmark cases for the Windmill AI generation modes (flow/app/script/cli/global) in ai_evals/. Use when adding or changing eval cases, or when running before/after benchmarks for AI chat / copilot changes.
+---
+
+# AI evals — authoring and running benchmark cases
+
+`ai_evals/` is a black-box benchmark runner for the Windmill AI generation modes:
+`flow`, `app`, `script`, `cli`, `global`. It always tests the **current** production
+prompts, tools, and guidance in this checkout. Each attempt runs the real production
+path, deterministic validation, then LLM judging.
+
+The goal is to test current production guidance with realistic user requests — **not**
+to pin one exact implementation shape.
+
+## Running benchmarks
+
+```bash
+cd ai_evals
+bun install                       # first time; frontend modes also need `cd frontend && bun install`
+bun run cli -- models             # list model aliases
+bun run cli -- cases global       # list cases for a mode
+bun run cli -- run global global-test1-script-create --model sonnet
+```
+
+Frontend modes (`flow`/`script`/`app`/`global`) route model calls through a Windmill
+backend's `/api/w/<ws>/ai/proxy`, so you need **any** reachable backend:
+
+```bash
+WMILL_AI_EVAL_BACKEND_URL=http://127.0.0.1:<port> WMILL_AI_EVAL_BACKEND_WORKSPACE=integration-tests \
+  bun run cli -- run global <caseIds...> --models sonnet,gpt-5.5,gemini-3.1-pro-preview
+```
+
+- **Reuse an existing workspace.** CE builds cap workspaces, so temp-workspace
+  creation 400s ("reached workspace limit"). Always set
+  `WMILL_AI_EVAL_BACKEND_WORKSPACE=integration-tests` (or any existing workspace) to
+  reuse one. The only side effect of a run is upserting an `f/evals/ai/<provider>`
+  resource there.
+- Provider keys live in `ai_evals/.env` and are auto-loaded by bun. The judge is a
+  separate Anthropic call (default `claude-sonnet-4-6`) regardless of the model under
+  test.
+
+## Authoring core rules
+
+1. Write prompts like a real user request.
+2. Prefer behavior, inputs, constraints, and outcomes over internal implementation.
+3. Keep deterministic validation narrow and hard.
+4. Put semantic expectations in `judgeChecklist`.
+5. Use `expected` fixtures only when exact structure really matters.
+
+### Prompt writing
+
+Prompts should sound like something a user would naturally ask. Do not write prompts
+as if the user knows Windmill internals unless the case explicitly tests a power-user
+workflow.
+
+Good:
+- "Create a flow that routes support requests based on customer tier."
+- "Add a reset button that sets the counter back to 0."
+- "Create a flow that reuses the existing greeting script instead of duplicating the logic."
+
+Bad:
+- "Use `branchone` with 3 branches and a default branch."
+- "Create a `rawscript` step with this exact topology."
+- "This is a benchmark harness."
+
+### Deterministic validation
+
+Use deterministic checks only for hard failures: missing required files; unexpected
+extra files when the prompt says not to create them; syntax errors; unresolved flow
+refs; missing required special modules or suspend config; obvious corruption.
+
+Do **not** encode one preferred implementation. Bad hard checks: exact step topology
+for a creation flow; exact branch structure when the prompt only asked for routing;
+exact input shape when multiple reasonable shapes are acceptable.
+
+### Judge checklist
+
+Every non-trivial case should have a `judgeChecklist` capturing user-visible behavior
+that must be present, important constraints, and key completion criteria — not
+low-level implementation details unless truly required.
+
+Good: "the flow calculates the order total with 8% tax"; "the flow reuses the existing
+workspace script instead of rewriting the logic". Bad: "uses `branchone`"; "contains a
+`rawscript` node".
+
+See `ai_evals/README.md` for the full case format, fields, and fixture details.
@@ -0,0 +1,50 @@
+---
+name: local-review-codex
+description: Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing. Same policy, model, and reasoning effort as the codex-pr-review GitHub action.
+---
+
+# Local Codex Review (pre-push)
+
+Runs the exact same review Codex performs in CI (`.github/workflows/codex-pr-review.yml`),
+but locally and scoped to work you have not pushed yet — so you catch what CI would flag
+before the PR exists. Use this before `git push` on a non-trivial change.
+
+**Correspondence with CI** — identical:
+- Policy: `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test coverage).
+- Model: `gpt-5.6-sol`, `model_reasoning_effort="xhigh"`.
+- Output: markdown starting with `## Codex Review`, findings tagged P0 / P1 / P2 with file:line.
+
+**Differences from CI** — local-only:
+- Scope is the current branch vs `main` at the merge-base, **including uncommitted changes** (CI reviews a pushed PR diff).
+- Sandbox is `read-only` (CI uses `danger-full-access` on an ephemeral runner). Codex reads the diff and files but cannot modify your working tree.
+- Fresh context is inherent: `codex exec` is a separate cold process, so it does not anchor on the current chat session — the same reason `local-review` insists on a subagent.
+
+## Prerequisites
+
+- `codex` CLI **>= 0.144.1** installed and authed (`codex login` or `OPENAI_API_KEY`). Older CLIs reject `gpt-5.6-sol` with "requires a newer version of Codex". Upgrade with `npm install --global @openai/codex@0.144.1` (may need `sudo` for a global install). Keep this in sync with the pin in `.github/workflows/codex-pr-review.yml`.
+- `git fetch` the base ref if it's stale, so the merge-base is accurate.
+
+## Run
+
+```bash
+bash .agents/skills/local-review-codex/run.sh          # review vs main (default)
+bash .agents/skills/local-review-codex/run.sh <base>   # review vs a different base ref
+```
+
+Invoke with `bash` (or run the executable directly) — the script needs Bash for
+`set -o pipefail`; `sh` is Dash on Debian/Ubuntu and would fail. If `main` isn't a
+local branch (e.g. a fresh single-branch checkout), the runner falls back to
+`origin/main` automatically.
+
+The script computes `BASE_SHA = git merge-base HEAD <base>`, feeds Codex `REVIEW.md` plus a
+diff context pointing at `git diff <BASE_SHA>` (which folds in uncommitted edits), and prints
+the review. It writes only temp files — nothing lands in the working tree.
+
+## Relaying the result
+
+Print the Codex output verbatim. Do not re-summarize or filter it — the value of a cold Codex
+pass is surfacing what the current session would rationalize away. Then decide with the user
+whether to address findings before pushing.
+
+For a Claude-native review instead, use `local-review` (branch-diff-reviewer subagent). This
+skill is the Codex counterpart; run both for independent perspectives.
@@ -0,0 +1,91 @@
+#!/usr/bin/env bash
+# Local Codex review — mirrors the .github/workflows/codex-pr-review.yml CI job,
+# but scoped to this branch's unpushed work (committed + uncommitted) so you can
+# review before pushing. Same policy (REVIEW.md), same model (gpt-5.6-sol) and
+# reasoning effort (xhigh) as CI. Runs read-only: Codex cannot modify your tree.
+#
+# Usage: run.sh [BASE_REF]   (BASE_REF defaults to "main")
+set -euo pipefail
+
+BASE_REF="${1:-main}"
+REPO_ROOT="$(git rev-parse --show-toplevel)"
+cd "$REPO_ROOT"
+
+if ! command -v codex >/dev/null 2>&1; then
+  echo "codex CLI not found. Install with: npm install --global @openai/codex@0.144.1" >&2
+  exit 1
+fi
+
+# Resolve the base to a concrete commit, preferring a local ref but falling back to
+# the remote-tracking ref — checkouts (CI, single-branch clones) often have only
+# origin/main, not a local main.
+if git rev-parse --verify --quiet "${BASE_REF}^{commit}" >/dev/null; then
+  BASE_COMMITISH="$BASE_REF"
+elif git rev-parse --verify --quiet "origin/${BASE_REF}^{commit}" >/dev/null; then
+  BASE_COMMITISH="origin/${BASE_REF}"
+else
+  echo "Base ref '$BASE_REF' not found as '$BASE_REF' or 'origin/$BASE_REF'. Try: git fetch origin $BASE_REF" >&2
+  exit 1
+fi
+
+# Diff from the merge-base so only this branch's changes are reviewed. Using the
+# base SHA with a single-ref `git diff` also folds in uncommitted working-tree edits,
+# but `git diff` never sees untracked files — those are gathered separately below so
+# brand-new files (a whole new module, a new skill dir) are not silently skipped.
+BASE_SHA="$(git merge-base HEAD "$BASE_COMMITISH")"
+HEAD_SHA="$(git rev-parse HEAD)"
+UNTRACKED="$(git ls-files --others --exclude-standard)"
+
+if [ "$BASE_SHA" = "$HEAD_SHA" ] && git diff --quiet "$BASE_SHA" && [ -z "$UNTRACKED" ]; then
+  echo "No changes vs $BASE_REF — nothing to review." >&2
+  exit 0
+fi
+
+PROMPT="$(mktemp)"
+OUT="$(mktemp)"
+trap 'rm -f "$PROMPT" "$OUT"' EXIT
+
+# REVIEW.md is the shared policy CI feeds Codex. Append the local output-format
+# and diff context inline (CI reads these from a generated context file; inlining
+# keeps the working tree clean — no scratch files land in the repo).
+cat REVIEW.md > "$PROMPT"
+cat >> "$PROMPT" <<EOF
+
+# Codex output format
+
+- This is a pre-push LOCAL review of unpushed work; there is no PR yet.
+- Inspect the changes by running the diff commands in the review context below.
+- Untracked files do NOT appear in \`git diff\`. Review every untracked path listed below by reading it directly (\`cat\`) — treat its entire contents as newly added.
+- Return markdown starting with \`## Codex Review\`.
+- Tag each finding with a severity (P0 / P1 / P2), file path, and line number when known confidently.
+
+# Review context
+
+Local review (pre-push): current branch vs $BASE_REF
+Base SHA: $BASE_SHA
+Head SHA: $HEAD_SHA (plus any uncommitted working-tree changes)
+
+Changed commits command:
+git log --oneline $BASE_SHA..HEAD
+
+Changed files command:
+git diff --stat $BASE_SHA
+
+Full review diff command (tracked changes, includes uncommitted edits):
+git diff --unified=0 $BASE_SHA
+
+Untracked files (NOT in the diff above — read each one directly, it is entirely new):
+$(if [ -n "$UNTRACKED" ]; then printf '%s\n' "$UNTRACKED"; else echo "(none)"; fi)
+EOF
+
+codex exec \
+  -C "$REPO_ROOT" \
+  -m gpt-5.6-sol \
+  -c 'model_reasoning_effort="xhigh"' \
+  -s read-only \
+  -o "$OUT" \
+  - < "$PROMPT"
+
+echo
+echo "===== Codex review ====="
+cat "$OUT"
@@ -1,18 +1,19 @@
 ---
 name: pr
 user_invocable: true
-description: Open a draft pull request on GitHub. MUST use when you want to create/open a PR.
+description: Open a draft pull request on GitHub and drive CI review rounds until it is ready. MUST use when you want to create/open a PR.
 ---

 # Pull Request Skill

-Create a draft pull request with a clear title and explicit description of changes.
+Create a draft pull request with a clear title and explicit description of changes, then drive it through CI review rounds to ready.

 ## Instructions

 1. **Analyze branch changes**: Understand all commits since diverging from main
 2. **Push to remote**: Ensure all commits are pushed
 3. **Create draft PR**: Always open as draft for review before merging
+4. **Drive review rounds**: trigger CI reviews on the draft and only flip to ready once every verdict is a go (see "Review rounds" below)

 ## PR Title Format

@@ -96,7 +97,11 @@ and continue once they confirm it's done.
 1. Run `git status` to check for uncommitted changes
 2. Run `git log main..HEAD --oneline` to see all commits in this branch
 3. Run `git diff main...HEAD` to see the full diff against main
-4. **Invoke the `local-review` skill** before creating the PR (`/local-review` in Claude Code, `$local-review` in Codex, `pi --skill local-review` / `/skill:local-review` in Pi). If issues are found, fix them and commit before proceeding. Do not skip this step.
+4. **Review the diff before creating the PR — run both reviews, do not skip:**
+   - **`local-review`** — Claude-native branch-diff-reviewer (`/local-review` in Claude Code, `$local-review` in Codex, `pi --skill local-review` / `/skill:local-review` in Pi).
+   - **`local-review-codex`** — cold Codex pass, the same review CI runs, for an independent perspective the Claude pass misses (`/local-review-codex` in Claude Code, or `bash .agents/skills/local-review-codex/run.sh`). If the `codex` CLI is missing or older than the version pinned in that skill, note it in your summary and continue — never block the PR on codex being unavailable.
+
+   Run both — they catch different things. If either surfaces issues, fix them and commit before proceeding.
 5. **Screenshots for frontend changes**: if `git diff main...HEAD --name-only` matches `^frontend/`, capture and embed screenshots of the affected UI per "Screenshots" above before writing the PR body (skip only if there is no visible UI effect).
 6. Check if remote branch exists and is up to date:
   ```bash
@@ -120,6 +125,40 @@ and continue once they confirm it's done.
   )"
   ```
 9. Return the PR URL to the user
+10. Drive the PR through CI review rounds to ready (see "Review rounds" below)
+
+## Review rounds (draft → ready)
+
+A PR leaves draft **only after a clean CI review round**. Never run `gh pr ready` before that.
+
+1. **Trigger a round and wait for it**: launch the waiter as a background Bash task (a round takes 10–30 min; you are woken when it exits — do not stop the session or poll in the foreground while it runs):
+
+   ```bash
+   bash .agents/skills/pr/review-round.sh <PR_NUMBER>
+   ```
+
+   It comments `/review` on the PR — which runs the Codex, Claude and Pi CI reviewers even on a draft — waits for the spawned `PR Review Commands` workflow run(s) to complete, then prints one verdict line per reviewer and saves the full review comments to files.
+
+2. **Judge the round.** Codex is mandatory; Claude, Pi and cubic count whenever they posted. Every review starts with one of the three `REVIEW.md` verdicts:
+   - Codex verdict missing → the round is void: comment `/codex` on the PR, wait for it the same way, and judge again.
+   - Any **"Should address issues before merging"** → fix the P0/P1 findings (and the nits while you're there), commit, push, and start a new round (step 1).
+   - Only **"Mergeable, but should ideally address nits"** and/or **"Good to merge"** → fix the nits too; a nit that is wrong or genuinely not worth fixing may instead be dismissed by replying to the review comment with your reasoning. Push nit-only fixes without starting another full round.
+
+3. **Flip to ready with the marker comment.** The review workflows skip the redundant `ready_for_review`-triggered round when the PR author has posted a marker naming the current head SHA **and** the PR's latest Codex review *posted before the marker* has a non-blocking verdict (reviewer evidence — a bare marker with no round behind it, or one whose last pre-marker Codex verdict is "Should address issues", skips nothing). Keep the prefix exact and use the full 40-char SHA of the head you are flipping:
+   - every verdict was "Good to merge" (head unchanged since the round):
+
+     `✅ Review round clean @ <head-sha>`
+
+   - nit-only round, nits fixed or dismissed afterwards (head may have moved past the reviewed SHA — say so):
+
+     `✅ Review round clean @ <head-sha> — nit-only verdicts at <round-sha>; nits addressed in <commit sha(s)> / dismissed in review replies`
+
+   ```bash
+   gh pr comment <PR_NUMBER> --body "✅ Review round clean @ $(git rev-parse HEAD)"
+   gh pr ready <PR_NUMBER>
+   ```
+
+   If any P0/P1 finding is unaddressed or the head moved for reasons other than nit fixes, do **not** post the marker or flip — run another round instead.

 ## EE Companion PR (when `*_ee.rs` files were modified)

@@ -0,0 +1,129 @@
+#!/usr/bin/env bash
+# Trigger a CI review round on a PR and wait for it to finish.
+#
+# Usage: bash .agents/skills/pr/review-round.sh [PR_NUMBER]
+#   PR_NUMBER defaults to the current branch's PR.
+#
+# Comments `/review` on the PR (works on drafts), waits for the spawned
+# "PR Review Commands" workflow run(s) to complete, then prints one verdict
+# line per reviewer and saves each full review comment to a file. A round
+# takes 10-30 minutes: run this in the background and act on its output when
+# it exits, per the pr skill ("Review rounds").
+set -euo pipefail
+
+REPO=${REPO:-$(gh repo view --json nameWithOwner --jq .nameWithOwner)}
+PR=${1:-$(gh pr view --json number --jq .number)}
+
+# Timestamp of the trigger comment, straight from GitHub, so local clock skew
+# can't make the run/comment filters below miss part of the round.
+TRIGGER_TIME=$(gh api "repos/$REPO/issues/$PR/comments" -f body='/review' --jq .created_at)
+echo "Review round triggered on $REPO#$PR at $TRIGGER_TIME"
+
+# Retry wrapper for one-off gh/API hiccups: a 45-minute wait must not die on
+# a single transient failure.
+retry() {
+  local attempt
+  for attempt in 1 2 3; do
+    if "$@"; then return 0; fi
+    sleep 10
+  done
+  return 1
+}
+
+# The /review comment spawns one "PR Review Commands" run holding the
+# claude/codex/pi jobs. Runs aren't linked to a PR, so wait on every run of
+# that workflow created after the trigger: a concurrent round on another PR
+# can only delay the answer, never truncate it. Every issue comment on any PR
+# spawns a fast-completing parse run of the same workflow, so the round's own
+# run may briefly lag the listing while unrelated runs already show completed:
+# require the all-completed state to hold past a floor and across two
+# consecutive polls before trusting it.
+DEADLINE=$(( $(date +%s) + 45 * 60 ))
+NO_RUN_DEADLINE=$(( $(date +%s) + 5 * 60 ))
+MIN_WAIT_UNTIL=$(( $(date +%s) + 3 * 60 ))
+STABLE=0
+FAILURES=0
+while :; do
+  if RUNS=$(gh run list --repo "$REPO" --workflow=pr-review-commands.yml \
+      --created ">=$TRIGGER_TIME" --limit 100 --json status); then
+    FAILURES=0
+  else
+    FAILURES=$(( FAILURES + 1 ))
+    if [ "$FAILURES" -ge 5 ]; then
+      echo "ERROR: listing workflow runs failed $FAILURES times in a row; aborting the wait." >&2
+      exit 1
+    fi
+    echo "WARNING: listing workflow runs failed (attempt $FAILURES/5); retrying in 60s." >&2
+    sleep 60
+    continue
+  fi
+  TOTAL=$(jq length <<<"$RUNS")
+  PENDING=$(jq '[.[] | select(.status != "completed")] | length' <<<"$RUNS")
+  NOW=$(date +%s)
+  if [ "$TOTAL" -gt 0 ] && [ "$PENDING" -eq 0 ] && [ "$NOW" -gt "$MIN_WAIT_UNTIL" ]; then
+    STABLE=$(( STABLE + 1 ))
+    if [ "$STABLE" -ge 2 ]; then
+      break
+    fi
+  else
+    STABLE=0
+  fi
+  if [ "$TOTAL" -eq 0 ] && [ "$NOW" -gt "$NO_RUN_DEADLINE" ]; then
+    echo "ERROR: no 'PR Review Commands' run appeared within 5 minutes of the /review comment; check that the comment author has write access and the workflow is enabled." >&2
+    exit 1
+  fi
+  if [ "$NOW" -gt "$DEADLINE" ]; then
+    echo "WARNING: review round still pending after 45 minutes; reporting whatever has been posted so far." >&2
+    break
+  fi
+  sleep 60
+done
+
+OUT_DIR=$(mktemp -d -t review-round-XXXXXX)
+COMMENTS_RAW=$(retry gh api "repos/$REPO/issues/$PR/comments?per_page=100" --paginate)
+jq -s --arg t "$TRIGGER_TIME" '[.[][] | select(.created_at > $t)]' \
+  <<<"$COMMENTS_RAW" > "$OUT_DIR/comments.json"
+# cubic posts through the PR reviews API, not issue comments.
+REVIEWS_RAW=$(retry gh api "repos/$REPO/pulls/$PR/reviews?per_page=100" --paginate)
+jq -s --arg t "$TRIGGER_TIME" '[.[][] | select((.submitted_at // "") > $t)]' \
+  <<<"$REVIEWS_RAW" > "$OUT_DIR/pr-reviews.json"
+
+VERDICT_RE='(Good to merge|Mergeable, but should ideally address nits|Should address issues before merging)'
+
+body_by_header() {
+  jq -r --arg h "$1" '[.[] | select(.body // "" | contains($h))] | last | .body // empty' \
+    "$OUT_DIR/comments.json"
+}
+body_by_login() {
+  jq -r --arg l "$1" '[.[] | select(.user.login == $l)] | last | .body // empty' \
+    "$OUT_DIR/comments.json"
+}
+report() { # <reviewer-name> <comment-body>
+  local name=$1 body=$2 verdict
+  if [ -z "$body" ]; then
+    echo "$name: (no review posted this round)"
+    return
+  fi
+  printf '%s\n' "$body" > "$OUT_DIR/$name.md"
+  verdict=$(printf '%s\n' "$body" | grep -m1 -oE "${VERDICT_RE}.*" | sed 's/\*\*//g' || true)
+  echo "$name: ${verdict:-(review posted but no verdict line; read $OUT_DIR/$name.md)}"
+}
+
+echo
+echo "=== Review round verdicts for $REPO#$PR (posted after $TRIGGER_TIME) ==="
+CODEX_BODY=$(body_by_header '## Codex Review')
+report codex "$CODEX_BODY"
+report claude "$(body_by_login 'claude[bot]')"
+report pi "$(body_by_header '## Pi Review')"
+CUBIC_BODY=$(jq -r '[.[] | select(.user.login | test("^cubic(-dev-ai)?(\\[bot\\])?$"; "i"))] | last | .body // empty' \
+  "$OUT_DIR/pr-reviews.json")
+if [ -z "$CUBIC_BODY" ]; then
+  CUBIC_BODY=$(jq -r '[.[] | select(.user.login | test("^cubic(-dev-ai)?(\\[bot\\])?$"; "i"))] | last | .body // empty' \
+    "$OUT_DIR/comments.json")
+fi
+report cubic "$CUBIC_BODY"
+echo
+echo "Full round output: $OUT_DIR (comments.json, pr-reviews.json, one .md per reviewer)"
+if [ -z "$CODEX_BODY" ]; then
+  echo "WARNING: Codex verdict missing - the round is incomplete. Re-trigger with a '/codex' PR comment and wait again." >&2
+fi
@@ -0,0 +1 @@
+../../../.agents/skills/ai-chat/SKILL.md
@@ -0,0 +1 @@
+../../../.agents/skills/ai-evals/SKILL.md
@@ -0,0 +1 @@
+../../../.agents/skills/local-review-codex/SKILL.md
@@ -28,7 +28,7 @@ ENV PATH="${PATH}:/usr/local/go/bin"
 ENV GO_PATH=/usr/local/go/bin/go

 # UV
-RUN curl --proto '=https' --tlsv1.2 -LsSf https://github.com/astral-sh/uv/releases/download/0.9.25/uv-installer.sh | sh && mv /usr/local/cargo/bin/uv /usr/local/bin/uv
+RUN curl --proto '=https' --tlsv1.2 -LsSf https://github.com/astral-sh/uv/releases/download/0.11.24/uv-installer.sh | sh && mv /usr/local/cargo/bin/uv /usr/local/bin/uv

 ENV TZ=Etc/UTC

@@ -1,5 +1,5 @@
 # Codex output format

- Read `./.github/codex/pr-review-context.md` for PR metadata and the diff commands.
+- Read the review context file whose absolute path is given at the end of these instructions; it holds the PR metadata and the diff commands.
 - Return a markdown PR comment starting with `## Codex Review`.
 - Tag each finding with a severity (P0 / P1 / P2), file path, and line number when known confidently.
@@ -1,6 +1,6 @@
 # Pi output format

- Read `./.github/pi/pr-review-context.md` for PR metadata and the diff commands.
+- Read the review context file whose absolute path is given at the end of these instructions; it holds the PR metadata and the diff (or the git commands to produce it).
 - Return a markdown PR comment starting with `## Pi Review`.
 - Tag each finding with a severity (P0 / P1 / P2), file path, and line number when known confidently.
 - Output ONLY the final review markdown — no preamble, no thinking, no tool transcripts.
@@ -0,0 +1,126 @@
+// Extracts every windmill.dev/docs link referenced in the frontend source and
+// verifies none of them 404. Run: `node .github/scripts/check-docs-links.mjs`.
+// Used by the check-docs-links GitHub workflow (release / manual trigger only).
+
+import { readdir, readFile } from 'node:fs/promises'
+import { join, extname } from 'node:path'
+
+const ROOT = 'frontend/src'
+const EXTS = new Set(['.ts', '.js', '.svelte', '.mjs', '.cjs'])
+const DOCS_RE = /https?:\/\/(?:www\.)?windmill\.dev\/docs\/[^\s"'`)>\]}]*/g
+// `const someBaseUrl = 'https://www.windmill.dev/docs/...'` used later as `${someBaseUrl}/foo`
+const BASE_RE = /(?:const|let|var)\s+([A-Za-z_$][\w$]*)\s*=\s*['"`](https?:\/\/(?:www\.)?windmill\.dev\/docs\/[^'"`]+)['"`]/g
+
+const CONCURRENCY = 24
+const TIMEOUT_MS = 20000
+const RETRIES = 2
+
+async function walk(dir) {
+	const out = []
+	for (const entry of await readdir(dir, { withFileTypes: true })) {
+		const p = join(dir, entry.name)
+		if (entry.isDirectory()) {
+			if (entry.name === 'node_modules' || entry.name === '.svelte-kit') continue
+			out.push(...(await walk(p)))
+		} else if (EXTS.has(extname(entry.name))) {
+			out.push(p)
+		}
+	}
+	return out
+}
+
+// url (no fragment) -> Set of source files it appears in
+const urls = new Map()
+const unresolved = []
+
+function record(url, file) {
+	const clean = url
+		.replace(/\\.*$/, '') // cut at an escape sequence embedded in a string literal (e.g. \n)
+		.replace(/#.*$/, '') // drop anchor fragment — irrelevant to a 404 check
+		.replace(/[.,;:'")\]]+$/, '')
+	if (!clean) return
+	// A `{`/`${` means the URL is built from an unresolved template/interpolation var.
+	if (clean.includes('{')) {
+		unresolved.push(`${clean}  (${file})`)
+		return
+	}
+	if (!urls.has(clean)) urls.set(clean, new Set())
+	urls.get(clean).add(file)
+}
+
+for (const file of await walk(ROOT)) {
+	let content = await readFile(file, 'utf8')
+	// Inline file-local base-url constants so `${base}/page` template literals resolve.
+	const bases = []
+	for (const m of content.matchAll(BASE_RE)) bases.push({ name: m[1], value: m[2], decl: m[0] })
+	for (const { name, value } of bases) {
+		content = content.replaceAll('${' + name + '}', value)
+	}
+	// Blank each base declaration so a prefix-only base (no index page of its own,
+	// e.g. .../app_configuration_settings) isn't checked as a standalone link.
+	// A genuinely bare `${base}` usage was already inlined above, so it's still covered.
+	for (const { decl } of bases) content = content.replace(decl, '')
+	for (const m of content.matchAll(DOCS_RE)) record(m[0], file)
+}
+
+const allUrls = [...urls.keys()].sort()
+console.log(`Found ${allUrls.length} distinct docs links across ${ROOT}`)
+if (unresolved.length) {
+	console.log(`\n⚠️  ${unresolved.length} link(s) built from an unrecognized base URL — skipped (register the base const so they can be checked):`)
+	for (const u of [...new Set(unresolved)].sort()) console.log(`   ${u}`)
+}
+
+async function check(url) {
+	for (let attempt = 0; attempt <= RETRIES; attempt++) {
+		const ctrl = new AbortController()
+		const timer = setTimeout(() => ctrl.abort(), TIMEOUT_MS)
+		try {
+			let res = await fetch(url, {
+				method: 'HEAD',
+				redirect: 'follow',
+				signal: ctrl.signal,
+				headers: { 'user-agent': 'windmill-docs-link-check' }
+			})
+			// Some hosts reject HEAD — fall back to GET.
+			if (res.status === 405 || res.status === 501) {
+				res = await fetch(url, {
+					method: 'GET',
+					redirect: 'follow',
+					signal: ctrl.signal,
+					headers: { 'user-agent': 'windmill-docs-link-check' }
+				})
+			}
+			clearTimeout(timer)
+			return { url, status: res.status, ok: res.status < 400 }
+		} catch (err) {
+			clearTimeout(timer)
+			if (attempt === RETRIES) return { url, status: 0, ok: false, error: String(err?.message || err) }
+			await new Promise((r) => setTimeout(r, 500 * (attempt + 1)))
+		}
+	}
+}
+
+// Simple concurrency pool.
+const results = []
+let idx = 0
+async function worker() {
+	while (idx < allUrls.length) {
+		const url = allUrls[idx++]
+		results.push(await check(url))
+	}
+}
+await Promise.all(Array.from({ length: CONCURRENCY }, worker))
+
+const failures = results.filter((r) => !r.ok)
+if (failures.length === 0) {
+	console.log(`\n✅ All ${allUrls.length} docs links are reachable.`)
+	process.exit(0)
+}
+
+console.log(`\n❌ ${failures.length} broken docs link(s):`)
+for (const f of failures.sort((a, b) => a.url.localeCompare(b.url))) {
+	console.log(`\n  ${f.url}`)
+	console.log(`    status: ${f.error ? `error (${f.error})` : f.status}`)
+	for (const file of urls.get(f.url)) console.log(`    ↳ ${file}`)
+}
+process.exit(1)
@@ -57,7 +57,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - uses: oven-sh/setup-bun@v2
        with:
@@ -69,7 +69,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - uses: oven-sh/setup-bun@v2
        with:
@@ -23,7 +23,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache: false
-          toolchain: 1.93.0
+          toolchain: 1.97.0
      - name: cargo check
        working-directory: ./backend
        timeout-minutes: 16
@@ -44,7 +44,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache: false
-          toolchain: 1.93.0
+          toolchain: 1.97.0
      - name: cargo check
        working-directory: ./backend
        timeout-minutes: 16
@@ -81,7 +81,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache: false
-          toolchain: 1.93.0
+          toolchain: 1.97.0
      - name: cargo check
        working-directory: ./backend
        timeout-minutes: 16
@@ -118,7 +118,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0
      - name: Fix stale v8 build cache
        working-directory: ./backend
        run: |
@@ -50,7 +50,12 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0
+          # This action defaults RUSTFLAGS to "-D warnings"; unset it so the test
+          # run is not failed by cross-platform dead-code (cfg(unix)-only helpers
+          # are unused on Windows). Warning hygiene is enforced on the Linux CI
+          # and the build_windows_worker_ release build, not this test job.
+          rustflags: ""

      - uses: actions/setup-dotnet@v4
        with:
@@ -58,7 +63,9 @@ jobs:

      - uses: denoland/setup-deno@v2
        with:
-          deno-version: v2.x
+          # Pin to the Deno version shipped in the runtime image (Dockerfile) so CI
+          # tests what production runs, instead of floating on the latest v2.x.
+          deno-version: 2.2.1

      - uses: actions/setup-go@v2
        with:
@@ -74,7 +81,7 @@ jobs:

      - uses: astral-sh/setup-uv@v6.2.1
        with:
-          version: "0.9.25"
+          version: "0.11.24"

      - uses: shivammathur/setup-php@v2
        with:
@@ -172,13 +179,17 @@ jobs:
          # binary link spikes several hundred MB of transient I/O. Capping at
          # 8 trades ~25% wall time for headroom on the ~75GB runner disk.
          CARGO_BUILD_JOBS: 8
-          # backend/Cargo.toml sets split-debuginfo = "unpacked", which on
-          # windows-msvc is coerced to "packed": every test-binary link spawns
-          # the mspdbsrv.exe PDB type server and writes a large .pdb. CI needs
-          # no debug info, so disable PDB generation for the dev/test profiles
-          # here (avoids both LNK1318 type-server limit and PDB disk usage).
-          CARGO_PROFILE_DEV_SPLIT_DEBUGINFO: "off"
-          CARGO_PROFILE_TEST_SPLIT_DEBUGINFO: "off"
+          # backend/Cargo.toml leaves profile.dev at the default debug = 2 for
+          # the (large) windmill workspace crates; that debuginfo is emitted
+          # into every object file and embedded in each test binary, and on
+          # windows-msvc also spawns the mspdbsrv.exe PDB type server. Across a
+          # full --all --features build it is the dominant consumer of the
+          # ~63GB free on the runner disk (LNK1180 / disk-full during linking).
+          # CI needs no debug info, so drop it entirely for the dev/test
+          # profiles here. debug = 0 supersedes the previous split-debuginfo=off
+          # knob (no debuginfo => no .pdb and no LNK1318 type-server limit).
+          CARGO_PROFILE_DEV_DEBUG: "0"
+          CARGO_PROFILE_TEST_DEBUG: "0"
          # Tests' poll-time stack frames (deep nested async fn chains in
          # debug builds) reach ~1.8MB. 4MB gives ~2x headroom against flaky
          # overflows under parallel-test contention.
@@ -197,9 +208,15 @@ jobs:
          WMDEBUG_FORCE_V0_WORKSPACE_DEPENDENCIES: 1
          WMDEBUG_FORCE_RUNNABLE_SETTINGS_V0: 1
          WMDEBUG_FORCE_NO_LEGACY_DEBOUNCING_COMPAT: 1
+        # Windows ships a worker-only binary, so test the crates a worker runs
+        # (windmill-worker/-common/-queue) via -p, not `--all`: this skips the
+        # disk-heavy windmill-api test binaries (LNK1180) and the server-only
+        # windmill-trigger-* crates (amqp does not build on Windows). Linux CI runs the rest.
        run: >
          cargo test
          --no-fail-fast
-          --features enterprise,deno_core,duckdb,license,python,rust,scoped_cache,parquet,private,csharp,php,quickjs,mcp,run_inline
-          --all
+          -p windmill-worker
+          -p windmill-common
+          -p windmill-queue
+          --features private,enterprise,deno_core,duckdb,python,rust,csharp,php,quickjs,parquet,mcp,scoped_cache,windmill-git-sync/private,windmill-object-store/private,windmill-object-store/enterprise
          -- --nocapture --test-threads=10
@@ -50,7 +50,9 @@ jobs:
          dotnet-version: "9.0.x"
      - uses: denoland/setup-deno@v2
        with:
-          deno-version: v2.x
+          # Pin to the Deno version shipped in the runtime image (Dockerfile) so CI
+          # tests what production runs, instead of floating on the latest v2.x.
+          deno-version: 2.2.1
      - uses: actions/setup-go@v2
        with:
          go-version: 1.21.5
@@ -62,7 +64,7 @@ jobs:
          node-version: "20"
      - uses: astral-sh/setup-uv@v6.2.1
        with:
-          version: "0.9.25"
+          version: "0.11.24"
      - uses: shivammathur/setup-php@v2
        with:
          php-version: "8.3"
@@ -88,7 +90,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0
      - name: Fix stale v8 build cache
        working-directory: ./backend
        run: |
@@ -237,13 +239,47 @@ jobs:
      - name: cargo test
        timeout-minutes: 30
        env:
+          # setup-rust-toolchain exports RUSTFLAGS=-D warnings, and the RUSTFLAGS env
+          # var fully REPLACES (never merges with) target.*.rustflags in
+          # backend/.cargo/config.toml. That silently drops the config's
+          # `-C link-arg=-fuse-ld=mold`, so CI links the many large integration-test
+          # binaries (v8 + duckdb + every language runtime, statically linked) with the
+          # default bfd linker. Its peak memory across ~12 parallel links OOM-kills the
+          # runner mid-link (SIGTERM => exit 143, before any test runs). Re-add the mold
+          # link arg here so CI links with mold like local dev, keeping -D warnings.
+          # (config.toml's `linker = "clang"` still applies; env only overrides rustflags.)
+          RUSTFLAGS: "-D warnings -C link-arg=-fuse-ld=mold"
          SQLX_OFFLINE: true
          DATABASE_URL: postgres://postgres:changeme@localhost:5432/windmill
          DISABLE_EMBEDDING: true
          RUST_LOG: "off"
          RUST_LOG_STYLE: never
          CARGO_NET_GIT_FETCH_WITH_CLI: true
-          CARGO_BUILD_JOBS: 12
+          # Cap parallel rustc/link jobs below the 16 available cores. The tail of
+          # the build links ~128 full-graph test binaries (one per tests/*.rs file
+          # across the workspace); at high parallelism enough heavy codegen+link
+          # units (rustc ~2.6GB, mold ~1GB each) overlap to exhaust the 64GB
+          # runner. Matches backend-test-windows.yml, which already uses 8.
+          CARGO_BUILD_JOBS: 8
+          # Incremental compilation is per-run dead weight in CI: rust-cache
+          # (cache-workspaces above) restores compiled dependency artifacts but
+          # never persists target/**/incremental, so there is no prior state to
+          # reuse in a one-shot `cargo test`. It only adds per-crate memory
+          # overhead and extra disk. Off here (kept on for local dev via
+          # .cargo/config.toml). Matches backend-test-windows.yml.
+          CARGO_INCREMENTAL: "0"
+          # backend/Cargo.toml leaves profile.dev at the default debug = 2 for
+          # the (large) windmill workspace crates; that debug info is emitted
+          # into every object file and embedded in each test binary. Across the
+          # full --all --features build it is the dominant memory/disk consumer
+          # when mold links the windmill-api-integration-tests binary, tipping
+          # the runner over (lost runner reported as a canceled step). CI needs
+          # no debug info, so drop it entirely for the dev/test profiles here.
+          # (test profile inherits dev, but the workspace crates link in as
+          # dev-profile deps, so both must be set.) CI-only; local dev builds
+          # are unaffected.
+          CARGO_PROFILE_DEV_DEBUG: "0"
+          CARGO_PROFILE_TEST_DEBUG: "0"
          # Tests' poll-time stack frames (deep nested async fn chains in
          # debug builds) reach ~1.8MB, leaving very thin headroom on the
          # default 2MB thread stack. 4MB gives ~2x buffer against flaky
@@ -5,6 +5,20 @@ env:
 name: Build caddy-l4
 on:
  workflow_dispatch:
+  push:
+    branches:
+      - main
+    paths:
+      - docker/DockerfileCaddyL4
+      - docker/entrypoint-caddy.sh
+      - docker/caddy-compat-normalize.awk
+      - docker/caddy-l4.version
+      - docker/test-caddy-compat.sh
+      - Caddyfile
+      # The version check below reads the pin out of docker-compose.yml, so a
+      # compose-only bump has to trigger this workflow or the check never runs.
+      - docker-compose.yml
+      - .github/workflows/build-caddy-l4-image.yml

 permissions: write-all

@@ -14,6 +28,35 @@ jobs:
    steps:
      - uses: actions/checkout@v4
      - uses: depot/setup-action@v1
+
+      # docker-compose.yml pins an exact tag, and the Caddyfile it must agree
+      # with lives in the same checkout. Fail the build rather than publish a
+      # version nothing references, which is how :latest drifted from the
+      # Caddyfile in the first place.
+      - name: Resolve and check image version
+        id: version
+        run: |
+          set -euo pipefail
+          version="$(tr -d '[:space:]' < docker/caddy-l4.version)"
+          pinned="$(grep -oE 'caddy-l4:[^[:space:]"]+' docker-compose.yml | head -1 | cut -d: -f2-)"
+          caddy="$(grep -m1 -oE '^FROM caddy:[0-9]+\.[0-9]+\.[0-9]+' docker/DockerfileCaddyL4 | cut -d: -f2)"
+          if [ "$version" != "$pinned" ]; then
+            echo "docker/caddy-l4.version is '$version' but docker-compose.yml pins '$pinned'" >&2
+            echo "Bump both together." >&2
+            exit 1
+          fi
+          # Otherwise a caddy bump that forgets the version file publishes a tag
+          # that names the wrong caddy.
+          case "$version" in
+            "$caddy"-*) ;;
+            *)
+              echo "docker/caddy-l4.version is '$version' but the Dockerfile pins caddy '$caddy'" >&2
+              echo "The version must be <caddy-version>-<revision>." >&2
+              exit 1
+              ;;
+          esac
+          echo "version=$version" >> "$GITHUB_OUTPUT"
+
      - name: Docker meta
        id: meta-ee-public
        uses: docker/metadata-action@v5
@@ -23,8 +66,22 @@ jobs:
          tags: |
            type=sha
            type=ref,event=branch
+            # Not gated on the default branch: docker-compose.yml pins this tag,
+            # so it has to be publishable from a branch (workflow_dispatch)
+            # before the pin merges, or main would reference a tag that does not
+            # exist yet. The version is immutable, so republishing from main is
+            # a no-op. Only branch pushes to main and manual dispatch run this
+            # workflow, so a branch cannot claim the tag by accident.
+            type=raw,value=${{ steps.version.outputs.version }}
            type=raw,value=latest,enable={{is_default_branch}}

+      # The shim rewrites config a self-hoster never sees, so a silent
+      # regression here strands them on a restart loop or a dead :80.
+      - name: Test the legacy-Caddyfile compatibility shim
+        run: |
+          docker build -f docker/DockerfileCaddyL4 -t caddy-l4:ci ./docker
+          docker/test-caddy-compat.sh caddy-l4:ci
+
      - name: Login to registry
        uses: docker/login-action@v3
        with:
@@ -33,7 +33,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - name: Substitute EE code
        shell: bash
@@ -45,8 +45,12 @@ jobs:
        env:
          RUSTFLAGS: "-D warnings"
        run: |
-          mkdir frontend/build && cd backend
+          cd backend
+          # Stub the openapi specs to empty: they are compiled in via an ungated
+          # include_str! but a worker binary never serves them, so this avoids
+          # embedding ~2.5MB of spec.
          New-Item -Path . -Name "windmill-api/openapi-deref.yaml" -ItemType "File" -Force
+          New-Item -Path . -Name "windmill-api/openapi-deref.json" -ItemType "File" -Force
          cargo check --features=ee_windows

      - name: Cargo build dynamic libraries windows
@@ -0,0 +1,23 @@
+name: Check frontend docs links
+
+on:
+  push:
+    tags:
+      - "v*"
+  workflow_dispatch:
+
+jobs:
+  check-docs-links:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          sparse-checkout: |
+            frontend/src
+            .github/scripts
+      - uses: actions/setup-node@v4
+        with:
+          node-version: "22.x"
+      - name: Verify docs links are not 404
+        run: node .github/scripts/check-docs-links.mjs
@@ -1,83 +0,0 @@
-name: Check Organization Membership
-
-on:
-  workflow_call:
-    inputs:
-      commenter:
-        required: false
-        type: string
-        default: ''
-        description: 'The username to check. Auto-detected from the event context if not provided.'
-      organization:
-        required: false
-        type: string
-        default: 'windmill-labs'
-        description: 'The organization to check membership for'
-      trusted_bot:
-        required: false
-        type: string
-        default: 'windmill-internal-app[bot]'
-        description: 'The trusted bot username to allow'
-    secrets:
-      access_token:
-        required: true
-        description: 'The access token to use for org membership check'
-    outputs:
-      is_member:
-        description: 'Whether the user is an organization member or trusted bot'
-        value: ${{ jobs.check-membership.outputs.is_member }}
-
-jobs:
-  check-membership:
-    runs-on: ubicloud-standard-2
-    outputs:
-      is_member: ${{ steps.check-membership.outputs.is_member }}
-    steps:
-      - name: Determine commenter
-        id: determine-commenter
-        run: |
-          COMMENTER="${{ inputs.commenter }}"
-          if [[ -z "$COMMENTER" ]]; then
-            if [[ "${{ github.event_name }}" == "issue_comment" || \
-                  "${{ github.event_name }}" == "pull_request_review_comment" ]]; then
-              COMMENTER="${{ github.event.comment.user.login }}"
-            elif [[ "${{ github.event_name }}" == "pull_request_review" ]]; then
-              COMMENTER="${{ github.event.review.user.login }}"
-            else
-              COMMENTER="${{ github.event.issue.user.login }}"
-            fi
-          fi
-          echo "commenter=$COMMENTER" >> $GITHUB_OUTPUT
-
-      - name: Check organization membership
-        id: check-membership
-        env:
-          ORG_ACCESS_TOKEN: ${{ secrets.access_token }}
-          COMMENTER: ${{ steps.determine-commenter.outputs.commenter }}
-          ORG: ${{ inputs.organization }}
-          TRUSTED_BOT: ${{ inputs.trusted_bot }}
-        run: |
-          # 1. Allow the trusted bot straight away
-          if [[ "$COMMENTER" == "$TRUSTED_BOT" ]]; then
-            echo "is_member=true" >> $GITHUB_OUTPUT
-            exit 0
-          fi
-
-          # 2. Disallow other bots
-          if [[ "${COMMENTER}" =~ \[bot\]$ ]]; then
-            echo "is_member=false" >> $GITHUB_OUTPUT
-            exit 0
-          fi
-
-          # 3. Otherwise check if the user is a member of the organization
-          STATUS=$(curl -s -o /dev/null -w "%{http_code}" \
-            -H "Authorization: token $ORG_ACCESS_TOKEN" \
-            -H "Accept: application/vnd.github+json" \
-            -H "X-GitHub-Api-Version: 2022-11-28" \
-            "https://api.github.com/orgs/$ORG/members/$COMMENTER")
-
-          if [ "$STATUS" -eq 204 ]; then
-            echo "is_member=true" >> $GITHUB_OUTPUT
-          else
-            echo "is_member=false" >> $GITHUB_OUTPUT
-          fi
@@ -0,0 +1,66 @@
+name: Check Write Access
+
+# Authorizes a user to trigger privileged command workflows (/review, /ai, /plan,
+# /updatesqlx, ...). The webhook author_association reports PRIVATE org members as
+# CONTRIBUTOR/NONE (only public members show as MEMBER), so command jobs can't gate on
+# it alone. This mints the internal GitHub App token — which can see private members —
+# and confirms the user is a member or has write access to the repo. The app token is
+# minted fresh per run, so unlike the old ORG_ACCESS_TOKEN PAT it never expires.
+
+on:
+  workflow_call:
+    inputs:
+      username:
+        required: true
+        type: string
+        description: 'The user whose access to verify'
+      trusted_bot:
+        required: false
+        type: string
+        default: 'windmill-internal-app[bot]'
+        description: 'A bot login that is always authorized'
+    outputs:
+      authorized:
+        description: 'true if the user is the trusted bot, an org member, or has repo write access'
+        value: ${{ jobs.check.outputs.authorized }}
+
+jobs:
+  check:
+    runs-on: ubuntu-latest
+    outputs:
+      authorized: ${{ steps.check.outputs.authorized }}
+    steps:
+      - name: Mint internal app token
+        id: app
+        uses: actions/create-github-app-token@v2
+        with:
+          app-id: ${{ vars.INTERNAL_APP_ID }}
+          private-key: ${{ secrets.INTERNAL_APP_KEY }}
+          owner: ${{ github.repository_owner }}
+
+      - name: Resolve authorization
+        id: check
+        env:
+          GH_TOKEN: ${{ steps.app.outputs.token }}
+          USERNAME: ${{ inputs.username }}
+          TRUSTED_BOT: ${{ inputs.trusted_bot }}
+          REPO: ${{ github.repository }}
+        run: |
+          if [ "$USERNAME" = "$TRUSTED_BOT" ]; then
+            echo "authorized=true" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+          ORG="${REPO%%/*}"
+          # Org membership resolves private members too (204 = member, 404 = not).
+          if gh api "orgs/$ORG/members/$USERNAME" --silent 2>/dev/null; then
+            echo "authorized=true" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+          # Fallback: effective repo permission (also covers outside collaborators).
+          PERM=$(gh api "repos/$REPO/collaborators/$USERNAME/permission" --jq '.permission' 2>/dev/null || echo none)
+          if [ "$PERM" = "admin" ] || [ "$PERM" = "write" ]; then
+            echo "authorized=true" >> "$GITHUB_OUTPUT"
+          else
+            echo "authorized=false" >> "$GITHUB_OUTPUT"
+            echo "$USERNAME is neither the trusted bot, an org member, nor a repo writer."
+          fi
@@ -11,20 +11,24 @@ on:
    types: [submitted]

 jobs:
-  check-membership:
+  # author_association misses private org members; check-access resolves them via the
+  # internal app token. Both are OR'd below so public members still pass instantly.
+  check-access:
    if: |
      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '/plan')) ||
      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '/plan')) ||
      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '/plan')) ||
      (github.event_name == 'issues' && contains(github.event.issue.body, '/plan'))
-    uses: ./.github/workflows/check-org-membership.yml
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
+    uses: ./.github/workflows/check-write-access.yml
+    with:
+      username: ${{ github.event.comment.user.login || github.event.review.user.login || github.event.issue.user.login }}
+    secrets: inherit

  claude-plan-action:
-    needs: check-membership
+    needs: [check-access]
    if: |
-      needs.check-membership.outputs.is_member == 'true'
+      needs.check-access.outputs.authorized == 'true' ||
+      contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association || github.event.review.author_association || github.event.issue.author_association)
    runs-on: ubicloud-standard-4
    timeout-minutes: 20
    permissions:
@@ -11,20 +11,24 @@ on:
    types: [submitted]

 jobs:
-  check-membership:
+  # author_association misses private org members; check-access resolves them via the
+  # internal app token. Both are OR'd below so public members still pass instantly.
+  check-access:
    if: |
      (github.event_name == 'issue_comment' && startsWith(github.event.comment.body, '/ai') && !startsWith(github.event.comment.body, '/ai-fast')) ||
      (github.event_name == 'pull_request_review_comment' && startsWith(github.event.comment.body, '/ai') && !startsWith(github.event.comment.body, '/ai-fast')) ||
      (github.event_name == 'pull_request_review' && startsWith(github.event.review.body, '/ai') && !startsWith(github.event.review.body, '/ai-fast')) ||
      (github.event_name == 'issues' && startsWith(github.event.issue.body, '/ai') && !startsWith(github.event.issue.body, '/ai-fast'))
-    uses: ./.github/workflows/check-org-membership.yml
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
+    uses: ./.github/workflows/check-write-access.yml
+    with:
+      username: ${{ github.event.comment.user.login || github.event.review.user.login || github.event.issue.user.login }}
+    secrets: inherit

  claude-code-action:
-    needs: check-membership
+    needs: [check-access]
    if: |
-      needs.check-membership.outputs.is_member == 'true'
+      needs.check-access.outputs.authorized == 'true' ||
+      contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association || github.event.review.author_association || github.event.issue.author_association)
    runs-on: ubicloud-standard-8
    permissions:
      contents: write
@@ -37,6 +41,44 @@ jobs:
        with:
          fetch-depth: 1

+      # Make the EE source (the *_ee.rs files in the companion repo) available so the
+      # reviewer can see EE-only code (e.g. windmill-queue/src/jobs_ee.rs), not just the
+      # CE surface. The EE ref is read from the PR head's backend/ee-repo-ref.txt (via the
+      # API, so it reflects the PR's EE pin regardless of which ref is checked out here).
+      - name: Check EE access
+        id: ee
+        env:
+          EE_TOKEN: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}
+          GH_TOKEN: ${{ github.token }}
+          PR_NUMBER: ${{ github.event.issue.number || github.event.pull_request.number }}
+        run: |
+          if [ -z "$EE_TOKEN" ] || [ -z "$PR_NUMBER" ]; then
+            echo "available=false" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+          HEAD_SHA=$(gh api "repos/${{ github.repository }}/pulls/$PR_NUMBER" --jq .head.sha)
+          REF=$(gh api "repos/${{ github.repository }}/contents/backend/ee-repo-ref.txt?ref=$HEAD_SHA" --jq .content | base64 -d | tr -d '[:space:]')
+          if [ -z "$REF" ]; then
+            echo "available=false" >> "$GITHUB_OUTPUT"
+          else
+            echo "available=true" >> "$GITHUB_OUTPUT"
+            echo "ee_repo_ref=$REF" >> "$GITHUB_OUTPUT"
+          fi
+
+      - name: Checkout EE repository
+        if: steps.ee.outputs.available == 'true'
+        uses: actions/checkout@v4
+        with:
+          repository: windmill-labs/windmill-ee-private
+          path: ./windmill-ee-private
+          ref: ${{ steps.ee.outputs.ee_repo_ref }}
+          token: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}
+          fetch-depth: 1
+
+      - name: Substitute EE code
+        if: steps.ee.outputs.available == 'true'
+        run: ./backend/substitute_ee_code.sh --copy --dir ./windmill-ee-private
+
      - name: Run Claude PR Action
        uses: anthropics/claude-code-action@v1
        with:
@@ -32,27 +32,19 @@ concurrency:
  cancel-in-progress: true

 jobs:
-  check-membership:
-    if: github.event_name == 'pull_request'
-    uses: ./.github/workflows/check-org-membership.yml
-    with:
-      commenter: ${{ github.event.pull_request.user.login }}
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
-
  codex-review:
-    needs: check-membership
    runs-on: ubicloud-standard-2
    timeout-minutes: 30
+    # A non-fork PR (head.repo.fork == false) can only be opened by someone with push
+    # access to this repo, so fork==false already enforces write access. Do NOT re-add
+    # an author_association gate: the pull_request webhook payload reports private org
+    # members as CONTRIBUTOR/NONE (only public members show as MEMBER), which silently
+    # skips auto-review for every private member.
    if: |
-      always() &&
+      github.event_name == 'workflow_call' ||
      (
-        needs.check-membership.result == 'skipped' ||
-        (needs.check-membership.result == 'success' && needs.check-membership.outputs.is_member == 'true')
-      ) &&
-      (
-        github.event_name == 'workflow_call' ||
-        (github.event.pull_request.draft == false && github.event.pull_request.head.repo.fork == false)
+        github.event.pull_request.draft == false &&
+        github.event.pull_request.head.repo.fork == false
      )
    permissions:
      contents: read
@@ -90,6 +82,7 @@ jobs:
          EVENT_BODY: ${{ github.event.pull_request.body }}
          EVENT_FORK: ${{ github.event.pull_request.head.repo.fork }}
          EVENT_AUTHOR: ${{ github.event.pull_request.user.login }}
+          EVENT_ACTION: ${{ github.event.action }}
        run: |
          if [ -n "$INPUT_PR_NUMBER" ]; then
            PR_JSON=$(gh pr view "$INPUT_PR_NUMBER" --repo "${{ github.repository }}" \
@@ -112,24 +105,68 @@ jobs:
            IS_FORK="$EVENT_FORK"
            PR_AUTHOR="$EVENT_AUTHOR"
          fi
-          if [ "$IS_FORK" = "true" ]; then
-            echo "Skipping Codex review for fork PR."
+          # Fork PRs run untrusted code with secrets present, so the automatic
+          # pull_request trigger never reviews them. A non-empty INPUT_PR_NUMBER
+          # means we arrived via workflow_call (a maintainer /codex comment gated
+          # by check-write-access), so allow forks only on that path.
+          if [ "$IS_FORK" = "true" ] && [ -z "$INPUT_PR_NUMBER" ]; then
+            echo "Skipping Codex review for fork PR (automatic trigger)."
            echo "skip=true" >> "$GITHUB_OUTPUT"
            exit 0
          fi
+          # An agent-driven PR flips to ready only after a clean /review round on
+          # a draft, marked by an author comment naming the head SHA (pr skill,
+          # "Review rounds"). Re-reviewing that same head on ready_for_review is
+          # redundant. The marker alone is author attestation, so also require
+          # reviewer evidence: a Codex review (posted by github-actions[bot], not
+          # forgeable by the author) that predates the marker and carries a
+          # non-blocking verdict. Comment-triggered and synchronize runs never
+          # skip. Keep the three copies of this check in sync (pr-ready-review /
+          # codex-pr-review / pi-pr-review); a shared local action would need the
+          # repo checked out before the check, which the fork paths here
+          # deliberately avoid.
+          if [ "$EVENT_ACTION" = "ready_for_review" ] && [ -z "$INPUT_PR_NUMBER" ]; then
+            # Fetch failures fail open (no skip): an API hiccup must run the
+            # review, never skip it or fail the job.
+            COMMENTS=$(gh api "repos/${{ github.repository }}/issues/$PR_NUMBER/comments?per_page=100" --paginate | jq -s '[.[][]]') || COMMENTS='[]'
+            MARKER_TIME=$(jq -r --arg author "$PR_AUTHOR" --arg marker "✅ Review round clean @ $HEAD_SHA" \
+              '[.[] | select(.user.login == $author) | select(.body | contains($marker)) | .created_at] | min // empty' <<<"$COMMENTS")
+            CODEX_VERDICT=''
+            if [ -n "$MARKER_TIME" ]; then
+              # Only Codex evidence that predates the marker counts: the ready-
+              # triggered Codex run itself posts after the flip and must not
+              # vouch for a sibling reviewer's skip.
+              CODEX_VERDICT=$(jq -r --arg mt "$MARKER_TIME" \
+                '[.[] | select(.user.login == "github-actions[bot]") | select(.body | contains("## Codex Review")) | select(.created_at < $mt)] | last | .body // ""' <<<"$COMMENTS" \
+                | grep -m1 -oE '(Good to merge|Mergeable, but should ideally address nits|Should address issues before merging)' || true)
+            fi
+            if [ -n "$MARKER_TIME" ] && [ -n "$CODEX_VERDICT" ] && [ "$CODEX_VERDICT" != "Should address issues before merging" ]; then
+              echo "Clean review round marker found for $HEAD_SHA with pre-marker non-blocking Codex verdict; skipping redundant review."
+              echo "skip=true" >> "$GITHUB_OUTPUT"
+              exit 0
+            fi
+          fi
+          # PR title/body are attacker-controlled free text. Use an unguessable
+          # per-run delimiter so a fork can't embed a fixed heredoc terminator to
+          # inject extra outputs — e.g. is_fork=false (last-write-wins), which
+          # would re-enable the EE checkout and trusted-path settings for forks.
+          RAND=$(head -c 16 /dev/urandom | od -An -tx1 | tr -d ' \n')
+          TITLE_EOF="TITLE_EOF_${RAND}"
+          BODY_EOF="BODY_EOF_${RAND}"
          {
            echo "skip=false"
+            echo "is_fork=$IS_FORK"
            echo "pr_number=$PR_NUMBER"
            echo "base_ref=$BASE_REF"
            echo "base_sha=$BASE_SHA"
            echo "head_sha=$HEAD_SHA"
            echo "pr_author=$PR_AUTHOR"
-            echo 'title<<PR_TITLE_EOF'
+            echo "title<<$TITLE_EOF"
            printf '%s\n' "$PR_TITLE"
-            echo 'PR_TITLE_EOF'
-            echo 'body<<PR_BODY_EOF'
+            echo "$TITLE_EOF"
+            echo "body<<$BODY_EOF"
            printf '%s\n' "$PR_BODY"
-            echo 'PR_BODY_EOF'
+            echo "$BODY_EOF"
          } >> "$GITHUB_OUTPUT"

      - name: Checkout repository
@@ -138,9 +175,17 @@ jobs:
        with:
          ref: refs/pull/${{ steps.pr.outputs.pr_number }}/merge
          fetch-depth: 1
+          # Don't persist github.token in .git/config: the review agent can read
+          # the checkout, and on the fork path that token (issue/PR write) would
+          # otherwise be exfiltratable. All later git ops target the public origin
+          # and need no auth; EE checkout and gh use their own explicit tokens.
+          persist-credentials: false

+      # Never expose the EE private-repo token to untrusted fork code. Skipping
+      # this step leaves steps.ee.outputs.available empty, so the EE checkout and
+      # substitution steps below are skipped too.
      - name: Check EE access
-        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
+        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true' && steps.pr.outputs.is_fork != 'true'
        id: ee
        env:
          EE_TOKEN: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}
@@ -174,7 +219,7 @@ jobs:

      - name: Install Codex CLI
        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
-        run: npm install --global @openai/codex@0.128.0
+        run: npm install --global @openai/codex@0.144.1

      - name: Configure Codex auth
        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
@@ -214,9 +259,12 @@ jobs:
          REPO: ${{ github.repository }}
          PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
        run: |
+          # Write outside the checkout: on the fork path the merge tree is
+          # attacker-controlled, and a committed symlink at this path would
+          # redirect the write.
          gh api "repos/$REPO/issues/$PR_NUMBER/comments?per_page=100" \
            --jq '[.[] | {user: .user.login, created_at: .created_at, body: (.body | .[:4000])}] | sort_by(.created_at) | .[-20:]' \
-            > prior-comments.json || echo "[]" > prior-comments.json
+            > "$RUNNER_TEMP/prior-comments.json" || echo "[]" > "$RUNNER_TEMP/prior-comments.json"

      - name: Write Codex review context
        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
@@ -230,9 +278,9 @@ jobs:
          PR_AUTHOR: ${{ steps.pr.outputs.pr_author }}
          EXTRA_PROMPT: ${{ inputs.extra_prompt }}
        run: |
-          mkdir -p .github/codex
          node <<'NODE'
          const fs = require('fs');
+          const tmp = process.env.RUNNER_TEMP;
          const lines = [
            `Repository: ${process.env.PR_REPOSITORY}`,
            `PR number: ${process.env.PR_NUMBER}`,
@@ -262,9 +310,9 @@ jobs:
          if (process.env.EXTRA_PROMPT && process.env.EXTRA_PROMPT.trim()) {
            lines.push('', 'Additional reviewer instructions:', process.env.EXTRA_PROMPT.trim());
          }
-          if (fs.existsSync('prior-comments.json')) {
+          if (fs.existsSync(`${tmp}/prior-comments.json`)) {
            try {
-              const comments = JSON.parse(fs.readFileSync('prior-comments.json', 'utf8'));
+              const comments = JSON.parse(fs.readFileSync(`${tmp}/prior-comments.json`, 'utf8'));
              if (Array.isArray(comments) && comments.length > 0) {
                lines.push(
                  '',
@@ -279,19 +327,39 @@ jobs:
              }
            } catch (_) {}
          }
-          fs.writeFileSync('.github/codex/pr-review-context.md', `${lines.join('\n')}\n`);
+          fs.writeFileSync(`${tmp}/pr-review-context.md`, `${lines.join('\n')}\n`);
          NODE

      - name: Run Codex review
        if: steps.codex_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
+        env:
+          PR_IS_FORK: ${{ steps.pr.outputs.is_fork }}
+          PR_BASE_REF: ${{ steps.pr.outputs.base_ref }}
        run: |
-          cat REVIEW.md .github/codex/pr-review.prompt.md > /tmp/codex-prompt.md
+          if [ "$PR_IS_FORK" = "true" ]; then
+            # Fork code is untrusted. Read the review policy/prompt from the base
+            # ref (git show) rather than the attacker-controlled merge checkout,
+            # so a malicious fork can't rewrite the reviewer's own instructions,
+            # and run in a network-disabled sandbox to block secret exfiltration.
+            git show "origin/$PR_BASE_REF:REVIEW.md" > /tmp/codex-prompt.md
+            git show "origin/$PR_BASE_REF:.github/codex/pr-review.prompt.md" >> /tmp/codex-prompt.md
+            SANDBOX_MODE=workspace-write
+          else
+            cat REVIEW.md .github/codex/pr-review.prompt.md > /tmp/codex-prompt.md
+            SANDBOX_MODE=danger-full-access
+          fi
+          # The context file lives in RUNNER_TEMP (outside the attacker-controlled
+          # checkout); tell the agent its absolute path.
+          printf '\nReview context file (absolute path): %s\n' "$RUNNER_TEMP/pr-review-context.md" >> /tmp/codex-prompt.md
+          # Write the final message outside the checkout too: a fork could commit
+          # codex-final-message.md as a symlink and redirect this write to overwrite
+          # e.g. a GitHub Action's index.js, which then runs with our credentials.
          codex exec \
            -C "$GITHUB_WORKSPACE" \
-            -m gpt-5.5 \
+            -m gpt-5.6-sol \
            -c 'model_reasoning_effort="xhigh"' \
-            -s danger-full-access \
-            -o codex-final-message.md \
+            -s "$SANDBOX_MODE" \
+            -o "$RUNNER_TEMP/codex-final-message.md" \
            - < /tmp/codex-prompt.md

      - name: Post Codex review comment
@@ -299,20 +367,52 @@ jobs:
        uses: actions/github-script@v7
        env:
          PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          CODEX_AUTH_JSON: ${{ secrets.CODEX_AUTH_JSON }}
+          GH_JOB_TOKEN: ${{ github.token }}
        with:
          github-token: ${{ github.token }}
          script: |
            const fs = require('fs');
-            const path = `${process.env.GITHUB_WORKSPACE}/codex-final-message.md`;
+            const path = `${process.env.RUNNER_TEMP}/codex-final-message.md`;
            if (!fs.existsSync(path)) {
              core.info('Codex did not produce a final message; skipping PR comment.');
              return;
            }
-            const body = fs.readFileSync(path, 'utf8').trim();
+            let body = fs.readFileSync(path, 'utf8').trim();
            if (!body) {
              core.info('Codex final message was empty; skipping PR comment.');
              return;
            }
+            // Defense-in-depth for fork reviews: the model call needs the provider
+            // credential in the env, and the posted comment bypasses Actions log
+            // masking. Strip any credential (API key, raw auth JSON, nested
+            // tokens) that leaked into the review text before posting.
+            const secrets = [];
+            const addSecret = (v, min) => {
+              if (typeof v === 'string' && v.length >= min) secrets.push(v);
+            };
+            addSecret(process.env.OPENAI_API_KEY, 8);
+            addSecret(process.env.CODEX_AUTH_JSON, 8);
+            addSecret(process.env.GH_JOB_TOKEN, 8);
+            if (process.env.CODEX_AUTH_JSON) {
+              try {
+                const collect = (o) => {
+                  if (typeof o === 'string') addSecret(o, 20);
+                  else if (Array.isArray(o)) o.forEach(collect);
+                  else if (o && typeof o === 'object') Object.values(o).forEach(collect);
+                };
+                collect(JSON.parse(process.env.CODEX_AUTH_JSON));
+              } catch (_) {}
+            }
+            for (const s of [...new Set(secrets)].sort((a, b) => b.length - a.length)) {
+              body = body.split(s).join('[REDACTED]');
+            }
+            body = body.trim();
+            if (!body) {
+              core.info('Codex final message was empty after redaction; skipping PR comment.');
+              return;
+            }
            await github.rest.issues.createComment({
              owner: context.repo.owner,
              repo: context.repo.repo,
@@ -5,21 +5,22 @@ on:
    types: [created]

 jobs:
-  check-membership:
-    if: >-
-      github.event.issue.pull_request && (
-        startsWith(github.event.comment.body, '/updatesqlx') ||
-        startsWith(github.event.comment.body, '/demo') ||
-        startsWith(github.event.comment.body, '/eeref') ||
-        startsWith(github.event.comment.body, '/docs')
-      )
-    uses: ./.github/workflows/check-org-membership.yml
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
+  # /command comments can come from anyone; author_association misses private org
+  # members, so check-access resolves them via the internal app token. Runs once and is
+  # OR'd into each job's guard (public members still pass on author_association alone).
+  check-access:
+    if: github.event.issue.pull_request != null && startsWith(github.event.comment.body, '/')
+    uses: ./.github/workflows/check-write-access.yml
+    with:
+      username: ${{ github.event.comment.user.login }}
+    secrets: inherit

  update-sqlx:
-    needs: check-membership
-    if: needs.check-membership.outputs.is_member == 'true' && startsWith(github.event.comment.body, '/updatesqlx')
+    needs: [check-access]
+    if: >-
+      github.event.issue.pull_request &&
+      (contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) || needs.check-access.outputs.authorized == 'true') &&
+      startsWith(github.event.comment.body, '/updatesqlx')
    runs-on: ubicloud-standard-8
    permissions:
      contents: write
@@ -79,7 +80,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - name: Install xmlsec and gssapi build-time deps
        run: |
@@ -147,8 +148,11 @@ jobs:
            })

  demo:
-    needs: check-membership
-    if: needs.check-membership.outputs.is_member == 'true' && startsWith(github.event.comment.body, '/demo')
+    needs: [check-access]
+    if: >-
+      github.event.issue.pull_request &&
+      (contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) || needs.check-access.outputs.authorized == 'true') &&
+      startsWith(github.event.comment.body, '/demo')
    runs-on: ubicloud-standard-2
    permissions:
      contents: read
@@ -227,8 +231,11 @@ jobs:
          fi

  update-ee-ref:
-    needs: check-membership
-    if: needs.check-membership.outputs.is_member == 'true' && startsWith(github.event.comment.body, '/eeref')
+    needs: [check-access]
+    if: >-
+      github.event.issue.pull_request &&
+      (contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) || needs.check-access.outputs.authorized == 'true') &&
+      startsWith(github.event.comment.body, '/eeref')
    runs-on: ubicloud-standard-2
    permissions:
      contents: write
@@ -313,8 +320,11 @@ jobs:
            })

  update-docs:
-    needs: check-membership
-    if: needs.check-membership.outputs.is_member == 'true' && startsWith(github.event.comment.body, '/docs')
+    needs: [check-access]
+    if: >-
+      github.event.issue.pull_request &&
+      (contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) || needs.check-access.outputs.authorized == 'true') &&
+      startsWith(github.event.comment.body, '/docs')
    runs-on: ubicloud-standard-2
    permissions:
      contents: read
@@ -9,6 +9,10 @@ on:
      - "backend/windmill-api-integration-tests/tests/git_sync*"
      - "backend/ee-repo-ref.txt"
      - "backend/windmill-common/src/workspaces.rs"
+      - "backend/windmill-worker/src/result_processor.rs"
+      - "backend/windmill-api-workspaces/**"
+      - "cli/src/commands/sync/**"
+      - "cli/src/utils/git.ts"
      - "integration_tests/test/git_sync_test.py"
      - ".github/workflows/git-sync-test.yml"
  pull_request:
@@ -18,6 +22,10 @@ on:
      - "backend/windmill-api-integration-tests/tests/git_sync*"
      - "backend/ee-repo-ref.txt"
      - "backend/windmill-common/src/workspaces.rs"
+      - "backend/windmill-worker/src/result_processor.rs"
+      - "backend/windmill-api-workspaces/**"
+      - "cli/src/commands/sync/**"
+      - "cli/src/utils/git.ts"
      - "integration_tests/test/git_sync_test.py"
      - ".github/workflows/git-sync-test.yml"

@@ -50,8 +58,8 @@ jobs:
          echo "Changed files:"
          echo "$CHANGED_FILES"

-          # Direct git sync file changes — always relevant
-          if echo "$CHANGED_FILES" | grep -qE '^(backend/windmill-git-sync/|backend/windmill-api-integration-tests/tests/git_sync|backend/windmill-common/src/workspaces\.rs|integration_tests/test/git_sync|\.github/workflows/git-sync-test\.yml)'; then
+          # Direct git sync file changes — always relevant.
+          if echo "$CHANGED_FILES" | grep -qE '^(backend/windmill-git-sync/|backend/windmill-worker/src/result_processor\.rs|backend/windmill-api-workspaces/|backend/windmill-api-integration-tests/tests/git_sync|backend/windmill-common/src/workspaces\.rs|cli/src/commands/sync/|cli/src/utils/git\.ts|integration_tests/test/git_sync|\.github/workflows/git-sync-test\.yml)'; then
            echo "should_run=true" >> "$GITHUB_OUTPUT"
            echo "Relevant: direct git sync file changes"
            exit 0
@@ -121,7 +129,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - uses: oven-sh/setup-bun@v2
        with:
@@ -180,6 +188,9 @@ jobs:
          DENO_PATH: deno
          BUN_PATH: bun
          NODE_BIN_PATH: node
+          # The auto-pull poller's SSRF guard rejects localhost git remotes;
+          # the tests' Gitea runs on localhost.
+          ALLOW_LOCAL_GIT_REMOTES: "true"
        run: |
          ./target/debug/windmill &
          echo "Waiting for Windmill to be ready..."
@@ -27,7 +27,7 @@ jobs:
          go build
      - name: Pushes to another repository
        id: push_directory
-        uses: cpina/github-action-push-to-another-repository@devel
+        uses: cpina/github-action-push-to-another-repository@55306faa4ed53b815ae49e564af8cfb359d32ae2 # v1.7.3
        env:
          API_TOKEN_GITHUB: ${{ secrets.DENO_PAT }}
        with:
@@ -30,27 +30,19 @@ concurrency:
  cancel-in-progress: true

 jobs:
-  check-membership:
-    if: github.event_name == 'pull_request'
-    uses: ./.github/workflows/check-org-membership.yml
-    with:
-      commenter: ${{ github.event.pull_request.user.login }}
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
-
  pi-review:
-    needs: check-membership
    runs-on: ubicloud-standard-2
    timeout-minutes: 30
+    # A non-fork PR (head.repo.fork == false) can only be opened by someone with push
+    # access to this repo, so fork==false already enforces write access. Do NOT re-add
+    # an author_association gate: the pull_request webhook payload reports private org
+    # members as CONTRIBUTOR/NONE (only public members show as MEMBER), which silently
+    # skips auto-review for every private member.
    if: |
-      always() &&
+      github.event_name == 'workflow_call' ||
      (
-        needs.check-membership.result == 'skipped' ||
-        (needs.check-membership.result == 'success' && needs.check-membership.outputs.is_member == 'true')
-      ) &&
-      (
-        github.event_name == 'workflow_call' ||
-        (github.event.pull_request.draft == false && github.event.pull_request.head.repo.fork == false)
+        github.event.pull_request.draft == false &&
+        github.event.pull_request.head.repo.fork == false
      )
    permissions:
      contents: read
@@ -83,6 +75,7 @@ jobs:
          EVENT_BODY: ${{ github.event.pull_request.body }}
          EVENT_FORK: ${{ github.event.pull_request.head.repo.fork }}
          EVENT_AUTHOR: ${{ github.event.pull_request.user.login }}
+          EVENT_ACTION: ${{ github.event.action }}
        run: |
          if [ -n "$INPUT_PR_NUMBER" ]; then
            PR_JSON=$(gh pr view "$INPUT_PR_NUMBER" --repo "${{ github.repository }}" \
@@ -105,24 +98,68 @@ jobs:
            IS_FORK="$EVENT_FORK"
            PR_AUTHOR="$EVENT_AUTHOR"
          fi
-          if [ "$IS_FORK" = "true" ]; then
-            echo "Skipping Pi review for fork PR."
+          # Fork PRs run untrusted code with secrets present, so the automatic
+          # pull_request trigger never reviews them. A non-empty INPUT_PR_NUMBER
+          # means we arrived via workflow_call (a maintainer /pi comment gated by
+          # check-write-access), so allow forks only on that path.
+          if [ "$IS_FORK" = "true" ] && [ -z "$INPUT_PR_NUMBER" ]; then
+            echo "Skipping Pi review for fork PR (automatic trigger)."
            echo "skip=true" >> "$GITHUB_OUTPUT"
            exit 0
          fi
+          # An agent-driven PR flips to ready only after a clean /review round on
+          # a draft, marked by an author comment naming the head SHA (pr skill,
+          # "Review rounds"). Re-reviewing that same head on ready_for_review is
+          # redundant. The marker alone is author attestation, so also require
+          # reviewer evidence: a Codex review (posted by github-actions[bot], not
+          # forgeable by the author) that predates the marker and carries a
+          # non-blocking verdict. Comment-triggered and synchronize runs never
+          # skip. Keep the three copies of this check in sync (pr-ready-review /
+          # codex-pr-review / pi-pr-review); a shared local action would need the
+          # repo checked out before the check, which the fork paths here
+          # deliberately avoid.
+          if [ "$EVENT_ACTION" = "ready_for_review" ] && [ -z "$INPUT_PR_NUMBER" ]; then
+            # Fetch failures fail open (no skip): an API hiccup must run the
+            # review, never skip it or fail the job.
+            COMMENTS=$(gh api "repos/${{ github.repository }}/issues/$PR_NUMBER/comments?per_page=100" --paginate | jq -s '[.[][]]') || COMMENTS='[]'
+            MARKER_TIME=$(jq -r --arg author "$PR_AUTHOR" --arg marker "✅ Review round clean @ $HEAD_SHA" \
+              '[.[] | select(.user.login == $author) | select(.body | contains($marker)) | .created_at] | min // empty' <<<"$COMMENTS")
+            CODEX_VERDICT=''
+            if [ -n "$MARKER_TIME" ]; then
+              # Only Codex evidence that predates the marker counts: the ready-
+              # triggered Codex run itself posts after the flip and must not
+              # vouch for a sibling reviewer's skip.
+              CODEX_VERDICT=$(jq -r --arg mt "$MARKER_TIME" \
+                '[.[] | select(.user.login == "github-actions[bot]") | select(.body | contains("## Codex Review")) | select(.created_at < $mt)] | last | .body // ""' <<<"$COMMENTS" \
+                | grep -m1 -oE '(Good to merge|Mergeable, but should ideally address nits|Should address issues before merging)' || true)
+            fi
+            if [ -n "$MARKER_TIME" ] && [ -n "$CODEX_VERDICT" ] && [ "$CODEX_VERDICT" != "Should address issues before merging" ]; then
+              echo "Clean review round marker found for $HEAD_SHA with pre-marker non-blocking Codex verdict; skipping redundant review."
+              echo "skip=true" >> "$GITHUB_OUTPUT"
+              exit 0
+            fi
+          fi
+          # PR title/body are attacker-controlled free text. Use an unguessable
+          # per-run delimiter so a fork can't embed a fixed heredoc terminator to
+          # inject extra outputs — e.g. is_fork=false (last-write-wins), which
+          # would re-enable the EE checkout and trusted-path settings for forks.
+          RAND=$(head -c 16 /dev/urandom | od -An -tx1 | tr -d ' \n')
+          TITLE_EOF="TITLE_EOF_${RAND}"
+          BODY_EOF="BODY_EOF_${RAND}"
          {
            echo "skip=false"
+            echo "is_fork=$IS_FORK"
            echo "pr_number=$PR_NUMBER"
            echo "base_ref=$BASE_REF"
            echo "base_sha=$BASE_SHA"
            echo "head_sha=$HEAD_SHA"
            echo "pr_author=$PR_AUTHOR"
-            echo 'title<<PR_TITLE_EOF'
+            echo "title<<$TITLE_EOF"
            printf '%s\n' "$PR_TITLE"
-            echo 'PR_TITLE_EOF'
-            echo 'body<<PR_BODY_EOF'
+            echo "$TITLE_EOF"
+            echo "body<<$BODY_EOF"
            printf '%s\n' "$PR_BODY"
-            echo 'PR_BODY_EOF'
+            echo "$BODY_EOF"
          } >> "$GITHUB_OUTPUT"

      - name: Checkout repository
@@ -131,9 +168,17 @@ jobs:
        with:
          ref: refs/pull/${{ steps.pr.outputs.pr_number }}/merge
          fetch-depth: 1
+          # Don't persist github.token in .git/config: the review agent can read
+          # the checkout, and on the fork path that token (issue/PR write) would
+          # otherwise be exfiltratable. All later git ops target the public origin
+          # and need no auth; EE checkout and gh use their own explicit tokens.
+          persist-credentials: false

+      # Never expose the EE private-repo token to untrusted fork code. Skipping
+      # this step leaves steps.ee.outputs.available empty, so the EE checkout and
+      # substitution steps below are skipped too.
      - name: Check EE access
-        if: steps.pi_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
+        if: steps.pi_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true' && steps.pr.outputs.is_fork != 'true'
        id: ee
        env:
          EE_TOKEN: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}
@@ -186,9 +231,12 @@ jobs:
          REPO: ${{ github.repository }}
          PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
        run: |
+          # Write outside the checkout: on the fork path the merge tree is
+          # attacker-controlled, and a committed symlink at this path would
+          # redirect the write.
          gh api "repos/$REPO/issues/$PR_NUMBER/comments?per_page=100" \
            --jq '[.[] | {user: .user.login, created_at: .created_at, body: (.body | .[:4000])}] | sort_by(.created_at) | .[-20:]' \
-            > prior-comments.json || echo "[]" > prior-comments.json
+            > "$RUNNER_TEMP/prior-comments.json" || echo "[]" > "$RUNNER_TEMP/prior-comments.json"

      - name: Write Pi review context
        if: steps.pi_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
@@ -202,9 +250,9 @@ jobs:
          PR_AUTHOR: ${{ steps.pr.outputs.pr_author }}
          EXTRA_PROMPT: ${{ inputs.extra_prompt }}
        run: |
-          mkdir -p .github/pi
          node <<'NODE'
          const fs = require('fs');
+          const tmp = process.env.RUNNER_TEMP;
          const lines = [
            `Repository: ${process.env.PR_REPOSITORY}`,
            `PR number: ${process.env.PR_NUMBER}`,
@@ -234,9 +282,9 @@ jobs:
          if (process.env.EXTRA_PROMPT && process.env.EXTRA_PROMPT.trim()) {
            lines.push('', 'Additional reviewer instructions:', process.env.EXTRA_PROMPT.trim());
          }
-          if (fs.existsSync('prior-comments.json')) {
+          if (fs.existsSync(`${tmp}/prior-comments.json`)) {
            try {
-              const comments = JSON.parse(fs.readFileSync('prior-comments.json', 'utf8'));
+              const comments = JSON.parse(fs.readFileSync(`${tmp}/prior-comments.json`, 'utf8'));
              if (Array.isArray(comments) && comments.length > 0) {
                lines.push(
                  '',
@@ -251,7 +299,7 @@ jobs:
              }
            } catch (_) {}
          }
-          fs.writeFileSync('.github/pi/pr-review-context.md', `${lines.join('\n')}\n`);
+          fs.writeFileSync(`${tmp}/pr-review-context.md`, `${lines.join('\n')}\n`);
          NODE

      - name: Run Pi review
@@ -259,16 +307,69 @@ jobs:
        env:
          DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
          PI_SKIP_VERSION_CHECK: '1'
+          PR_IS_FORK: ${{ steps.pr.outputs.is_fork }}
+          PR_BASE_REF: ${{ steps.pr.outputs.base_ref }}
+          PR_BASE_SHA: ${{ steps.pr.outputs.base_sha }}
+          PR_HEAD_SHA: ${{ steps.pr.outputs.head_sha }}
        run: |
          set -o pipefail
-          cat REVIEW.md .github/pi/pr-review.prompt.md > /tmp/pi-prompt.md
+          PI_HARDEN_FLAGS=()
+          # Keep generated files (final message, events, context) outside the
+          # checkout: on the fork path a committed symlink at any of these paths
+          # would redirect our write and could overwrite an action's code that
+          # then runs with our credentials. RUNNER_TEMP is outside the checkout.
+          OUT_DIR="$RUNNER_TEMP"
+          CTX="$RUNNER_TEMP/pr-review-context.md"
+          if [ "$PR_IS_FORK" = "true" ]; then
+            # Fork code is untrusted. Read the review policy/prompt from the base
+            # ref (git show) rather than the attacker-controlled merge checkout,
+            # so a malicious fork can't rewrite the reviewer's own instructions,
+            # and drop the bash tool so the agent has no shell to exfiltrate with.
+            git show "origin/$PR_BASE_REF:REVIEW.md" > /tmp/pi-prompt.md
+            git show "origin/$PR_BASE_REF:.github/pi/pr-review.prompt.md" >> /tmp/pi-prompt.md
+            PI_TOOLS=read,grep,find,ls
+
+            # The agent has no shell, so pre-compute the diff (base...head SHAs are
+            # trusted) into the context file it reads. It may still read fork files
+            # by absolute path for extra context — reads are safe.
+            {
+              echo ""
+              echo "## Pre-computed review diff (base...head)"
+              echo "You have no shell. The full diff is below. The repository checkout"
+              echo "is at $GITHUB_WORKSPACE — you may read files there by absolute path."
+              echo '```diff'
+              git -C "$GITHUB_WORKSPACE" diff --unified=0 "$PR_BASE_SHA...$PR_HEAD_SHA"
+              echo '```'
+            } >> "$CTX"
+
+            # Pi resolves ALL project config from <cwd>/.pi (settings/packages,
+            # extensions, skills, themes, prompts, SYSTEM.md); inside the fork
+            # checkout a fork could inject any to run code or rewrite our system
+            # prompt. Discovery is cwd-based, so run from a fresh empty dir.
+            PI_WORKDIR=$(mktemp -d)
+            cd "$PI_WORKDIR"
+
+            # Belt-and-suspenders on top of the isolated cwd: refuse discovery of
+            # extensions/skills/templates/themes/context-files, and PI_OFFLINE=1 to
+            # block any startup network op or package install. PI_OFFLINE gates only
+            # startup network ops, not the provider inference call.
+            PI_HARDEN_FLAGS=(--no-extensions --no-skills --no-prompt-templates --no-themes --no-context-files)
+            export PI_OFFLINE=1
+          else
+            cat REVIEW.md .github/pi/pr-review.prompt.md > /tmp/pi-prompt.md
+            PI_TOOLS=read,grep,find,ls,bash
+          fi
+          # The context file lives in RUNNER_TEMP (outside the checkout); tell the
+          # agent its absolute path.
+          printf '\nReview context file (absolute path): %s\n' "$CTX" >> /tmp/pi-prompt.md
          pi -p \
            --provider deepseek \
            --model deepseek-v4-pro \
-            --tools read,grep,find,ls,bash \
+            --tools "$PI_TOOLS" \
+            "${PI_HARDEN_FLAGS[@]}" \
            --mode json \
            < /tmp/pi-prompt.md \
-            | tee pi-events.jsonl \
+            | tee "$OUT_DIR/pi-events.jsonl" \
            | jq -rc --unbuffered '
                if .type == "agent_start" then "🤖 pi agent started"
                elif .type == "turn_start" then "── turn ──"
@@ -296,27 +397,43 @@ jobs:
            | map(select(.role == "assistant"))
            | last
            | (.content[]? | select(.type == "text") | .text)
-          ' pi-events.jsonl > pi-final-message.md
+          ' "$OUT_DIR/pi-events.jsonl" > "$OUT_DIR/pi-final-message.md"

      - name: Post Pi review comment
        if: steps.pi_config.outputs.enabled == 'true' && steps.pr.outputs.skip != 'true'
        uses: actions/github-script@v7
        env:
          PR_NUMBER: ${{ steps.pr.outputs.pr_number }}
+          DEEPSEEK_API_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
+          GH_JOB_TOKEN: ${{ github.token }}
        with:
          github-token: ${{ github.token }}
          script: |
            const fs = require('fs');
-            const path = `${process.env.GITHUB_WORKSPACE}/pi-final-message.md`;
+            const path = `${process.env.RUNNER_TEMP}/pi-final-message.md`;
            if (!fs.existsSync(path)) {
              core.info('Pi did not produce a final message; skipping PR comment.');
              return;
            }
-            const body = fs.readFileSync(path, 'utf8').trim();
+            let body = fs.readFileSync(path, 'utf8').trim();
            if (!body) {
              core.info('Pi final message was empty; skipping PR comment.');
              return;
            }
+            // Defense-in-depth for fork reviews: the model call needs the provider
+            // credential in the environment (readable via /proc/self/environ), and
+            // the posted comment is an exfiltration channel that bypasses GitHub
+            // Actions log masking. Strip the credential if it leaked into the text.
+            for (const s of [process.env.DEEPSEEK_API_KEY, process.env.GH_JOB_TOKEN]) {
+              if (typeof s === 'string' && s.length >= 8) {
+                body = body.split(s).join('[REDACTED]');
+              }
+            }
+            body = body.trim();
+            if (!body) {
+              core.info('Pi final message was empty after redaction; skipping PR comment.');
+              return;
+            }
            await github.rest.issues.createComment({
              owner: context.repo.owner,
              repo: context.repo.repo,
@@ -30,38 +30,73 @@ concurrency:
  cancel-in-progress: true

 jobs:
-  check-membership:
-    if: github.event_name == 'pull_request'
-    uses: ./.github/workflows/check-org-membership.yml
-    with:
-      commenter: ${{ github.event.pull_request.user.login }}
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
-
  auto-review:
-    needs: check-membership
    runs-on: ubuntu-latest
+    # A non-fork PR (head.repo.fork == false) can only be opened by someone with push
+    # access to this repo, so fork==false already enforces write access. Do NOT re-add
+    # an author_association gate: the pull_request webhook payload reports private org
+    # members as CONTRIBUTOR/NONE (only public members show as MEMBER), which silently
+    # skips auto-review for every private member.
    if: |
-      always() &&
+      github.event_name == 'workflow_call' ||
      (
-        needs.check-membership.result == 'skipped' ||
-        (needs.check-membership.result == 'success' && needs.check-membership.outputs.is_member == 'true')
-      ) &&
-      (
-        github.event_name == 'workflow_call' ||
-        (github.event.pull_request.draft == false || github.event.pull_request.ready_for_review == true)
+        (github.event.pull_request.draft == false || github.event.pull_request.ready_for_review == true) &&
+        github.event.pull_request.head.repo.fork == false
      )
    permissions:
      contents: read
      pull-requests: read
      id-token: write
    steps:
+      # An agent-driven PR flips to ready only after a clean /review round on a
+      # draft, marked by an author comment naming the head SHA (pr skill, "Review
+      # rounds"). Re-reviewing that same head on ready_for_review is redundant.
+      # The marker alone is author attestation, so also require reviewer evidence:
+      # a Codex review (posted by github-actions[bot], not forgeable by the author)
+      # that predates the marker and carries a non-blocking verdict. Comment-
+      # triggered (workflow_call) and opened runs never skip. Keep the three
+      # copies of this check in sync (pr-ready-review / codex-pr-review /
+      # pi-pr-review); a shared local action would need the repo checked out
+      # before the check, which the codex/pi fork paths deliberately avoid.
+      - name: Check clean-round marker
+        id: marker
+        if: github.event_name == 'pull_request' && github.event.action == 'ready_for_review'
+        env:
+          GH_TOKEN: ${{ github.token }}
+          REPO: ${{ github.repository }}
+          PR_NUMBER: ${{ github.event.pull_request.number }}
+          HEAD_SHA: ${{ github.event.pull_request.head.sha }}
+          PR_AUTHOR: ${{ github.event.pull_request.user.login }}
+        run: |
+          # Fetch failures fail open (skip=false): an API hiccup must run the
+          # review, never skip it or fail the job.
+          COMMENTS=$(gh api "repos/$REPO/issues/$PR_NUMBER/comments?per_page=100" --paginate | jq -s '[.[][]]') || COMMENTS='[]'
+          MARKER_TIME=$(jq -r --arg author "$PR_AUTHOR" --arg marker "✅ Review round clean @ $HEAD_SHA" \
+            '[.[] | select(.user.login == $author) | select(.body | contains($marker)) | .created_at] | min // empty' <<<"$COMMENTS")
+          CODEX_VERDICT=''
+          if [ -n "$MARKER_TIME" ]; then
+            # Only Codex evidence that predates the marker counts: the ready-
+            # triggered Codex run itself posts after the flip and must not vouch
+            # for a sibling reviewer's skip.
+            CODEX_VERDICT=$(jq -r --arg mt "$MARKER_TIME" \
+              '[.[] | select(.user.login == "github-actions[bot]") | select(.body | contains("## Codex Review")) | select(.created_at < $mt)] | last | .body // ""' <<<"$COMMENTS" \
+              | grep -m1 -oE '(Good to merge|Mergeable, but should ideally address nits|Should address issues before merging)' || true)
+          fi
+          if [ -n "$MARKER_TIME" ] && [ -n "$CODEX_VERDICT" ] && [ "$CODEX_VERDICT" != "Should address issues before merging" ]; then
+            echo "Clean review round marker found for $HEAD_SHA with pre-marker non-blocking Codex verdict; skipping redundant review."
+            echo "skip=true" >> "$GITHUB_OUTPUT"
+          else
+            echo "skip=false" >> "$GITHUB_OUTPUT"
+          fi
+
      - name: Checkout repository
+        if: steps.marker.outputs.skip != 'true'
        uses: actions/checkout@v5
        with:
          fetch-depth: 1

      - name: Check EE access
+        if: steps.marker.outputs.skip != 'true'
        id: ee
        env:
          EE_TOKEN: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}
@@ -88,6 +123,7 @@ jobs:
        run: ./backend/substitute_ee_code.sh --copy --dir ./windmill-ee-private

      - name: Resolve PR number
+        if: steps.marker.outputs.skip != 'true'
        id: resolve
        env:
          GH_TOKEN: ${{ github.token }}
@@ -107,6 +143,7 @@ jobs:
          echo "pr_author=$PR_AUTHOR" >> "$GITHUB_OUTPUT"

      - name: Fetch prior PR discussion
+        if: steps.marker.outputs.skip != 'true'
        id: prior
        env:
          GH_TOKEN: ${{ github.token }}
@@ -125,6 +162,7 @@ jobs:
          ' prior-comments.json > prior-comments.md

      - name: Read review prompt
+        if: steps.marker.outputs.skip != 'true'
        id: review-prompt
        env:
          EXTRA_PROMPT: ${{ inputs.extra_prompt }}
@@ -148,6 +186,7 @@ jobs:
          } >> "$GITHUB_ENV"

      - name: Automatic PR Review
+        if: steps.marker.outputs.skip != 'true'
        uses: anthropics/claude-code-action@v1
        with:
          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
@@ -42,16 +42,24 @@ jobs:
              ;;
          esac

-  check-membership:
-    needs: parse
+  # author_association misses private org members; check-access resolves them via the
+  # internal app token. Both are OR'd so public members still pass instantly.
+  check-access:
+    needs: [parse]
    if: needs.parse.outputs.command != ''
-    uses: ./.github/workflows/check-org-membership.yml
-    secrets:
-      access_token: ${{ secrets.ORG_ACCESS_TOKEN }}
+    uses: ./.github/workflows/check-write-access.yml
+    with:
+      username: ${{ github.event.comment.user.login }}
+    secrets: inherit

  acknowledge:
-    needs: [parse, check-membership]
-    if: needs.parse.outputs.command != '' && needs.check-membership.outputs.is_member == 'true'
+    needs: [parse, check-access]
+    if: |
+      needs.parse.outputs.command != '' &&
+      (
+        contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) ||
+        needs.check-access.outputs.authorized == 'true'
+      )
    runs-on: ubuntu-latest
    permissions:
      issues: write
@@ -68,9 +76,12 @@ jobs:
            -f content=eyes >/dev/null

  claude:
-    needs: [parse, check-membership]
+    needs: [parse, check-access]
    if: |
-      needs.check-membership.outputs.is_member == 'true' &&
+      (
+        contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) ||
+        needs.check-access.outputs.authorized == 'true'
+      ) &&
      (needs.parse.outputs.command == 'review' || needs.parse.outputs.command == 'claude')
    permissions:
      contents: read
@@ -86,9 +97,12 @@ jobs:
      WINDMILL_EE_PRIVATE_ACCESS: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}

  codex:
-    needs: [parse, check-membership]
+    needs: [parse, check-access]
    if: |
-      needs.check-membership.outputs.is_member == 'true' &&
+      (
+        contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) ||
+        needs.check-access.outputs.authorized == 'true'
+      ) &&
      (needs.parse.outputs.command == 'review' || needs.parse.outputs.command == 'codex')
    permissions:
      contents: read
@@ -105,9 +119,12 @@ jobs:
      WINDMILL_EE_PRIVATE_ACCESS: ${{ secrets.WINDMILL_EE_PRIVATE_ACCESS }}

  pi:
-    needs: [parse, check-membership]
+    needs: [parse, check-access]
    if: |
-      needs.check-membership.outputs.is_member == 'true' &&
+      (
+        contains(fromJSON('["OWNER", "MEMBER", "COLLABORATOR"]'), github.event.comment.author_association) ||
+        needs.check-access.outputs.authorized == 'true'
+      ) &&
      (needs.parse.outputs.command == 'review' || needs.parse.outputs.command == 'pi')
    permissions:
      contents: read
@@ -35,7 +35,7 @@ jobs:
      - uses: actions-rust-lang/setup-rust-toolchain@v1
        with:
          cache-workspaces: backend
-          toolchain: 1.93.0
+          toolchain: 1.97.0

      - name: Substitute EE code
        shell: bash
@@ -56,8 +56,12 @@ jobs:
          vcpkg.exe integrate install
          $env:VCPKGRS_DYNAMIC=1
          $env:OPENSSL_DIR="${Env:VCPKG_INSTALLATION_ROOT}\installed\x64-windows-static"
-          mkdir frontend/build && cd backend
+          cd backend
+          # Stub the openapi specs to empty: they are compiled in via an ungated
+          # include_str! but a worker binary never serves them, so this avoids
+          # embedding ~2.5MB of spec.
          New-Item -Path . -Name "windmill-api/openapi-deref.yaml" -ItemType "File" -Force
+          New-Item -Path . -Name "windmill-api/openapi-deref.json" -ItemType "File" -Force
          cargo build --release --features=ee_windows
      - name: Rename binary with corresponding architecture
        run: |
@@ -0,0 +1,52 @@
+name: Refresh docs snapshot
+
+# The backend embeds a vendored docs snapshot (backend/windmill-api/docs_snapshot/*.gz)
+# so in-product docs search works with no runtime egress. This job re-fetches it from
+# windmill.dev on a schedule and opens a PR when it changed, keeping the embedded docs
+# fresh independently of the release cadence (the binary embeds whatever is on the
+# source tree at build time, so a merged refresh rides into the next release build).
+on:
+  schedule:
+    - cron: "0 6 * * 1" # Mondays 06:00 UTC
+  workflow_dispatch:
+
+jobs:
+  refresh:
+    runs-on: ubicloud
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      - uses: actions/create-github-app-token@v2
+        id: app
+        with:
+          app-id: ${{ vars.INTERNAL_APP_ID }}
+          private-key: ${{ secrets.INTERNAL_APP_KEY }}
+      - uses: actions/checkout@v4
+        with:
+          token: ${{ steps.app.outputs.token }}
+      - name: Fetch + re-gzip docs snapshot
+        run: cd backend/windmill-api/docs_snapshot && ./fetch.sh
+      - name: Sanity-check the fetched corpus
+        # curl -f in fetch.sh rejects HTTP errors, but not a valid-but-garbage 200
+        # (truncated file, error page). Guard against embedding a broken snapshot.
+        run: |
+          cd backend/windmill-api/docs_snapshot
+          test "$(wc -c < llms-full.txt.gz)" -gt 100000
+          test "$(wc -c < llms.txt.gz)" -gt 1000
+          pages=$(gzip -dc llms-full.txt.gz | grep -c '^Source:' || true)
+          echo "pages in snapshot: $pages"
+          test "${pages:-0}" -ge 200
+      - uses: peter-evans/create-pull-request@v6
+        with:
+          token: ${{ steps.app.outputs.token }}
+          branch: chore/refresh-docs-snapshot
+          add-paths: backend/windmill-api/docs_snapshot/*.gz
+          commit-message: "chore: refresh vendored docs snapshot"
+          title: "chore: refresh vendored docs snapshot"
+          body: |
+            Automated refresh of the embedded docs snapshot
+            (`backend/windmill-api/docs_snapshot/*.gz`) from windmill.dev.
+
+            Review the diff for unexpected churn (a bad upstream docs deploy would
+            show up as a large drop in pages or content) before merging.
@@ -16,7 +16,7 @@ Open-source platform for internal tools, workflows, API integrations, background
 - **Backend patterns**: use the `rust-backend` skill when writing Rust code
 - **Frontend patterns**: use the `svelte-frontend` skill when writing Svelte code. Do NOT edit svelte files unless you have read that skill.
 - **Frontend UUIDs**: do not call `crypto.randomUUID()` in frontend code. Import `randomUUID` from `$lib/utils/uuid` instead.
- **Code review**: review the current PR or branch against the shared review policy in `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test-coverage assessment). The skill at `.agents/skills/local-review/SKILL.md` orchestrates it. All three CLIs auto-discover the same SKILL — Claude reads `.claude/skills/` (symlinked to the canonical `.agents/skills/` file), Codex and Pi read `.agents/skills/` directly. Invoke with `/local-review` in Claude Code, `$local-review` (or `/skills` selector) in Codex, or `pi --skill local-review` / `/skill:local-review` in Pi.
+- **Code review**: review the current PR or branch against the shared review policy in `REVIEW.md` (severity triage, public-surface checklist, AGENTS.md compliance, test-coverage assessment). The skill at `.agents/skills/local-review/SKILL.md` orchestrates it. All three CLIs auto-discover the same SKILL — Claude reads `.claude/skills/` (symlinked to the canonical `.agents/skills/` file), Codex and Pi read `.agents/skills/` directly. Invoke with `/local-review` in Claude Code, `$local-review` (or `/skills` selector) in Codex, or `pi --skill local-review` / `/skill:local-review` in Pi. For a Codex-driven pass that mirrors the `codex-pr-review` GitHub action against your unpushed work (committed + uncommitted) before you push, use `/local-review-codex` (`.agents/skills/local-review-codex/`) — same `REVIEW.md` policy, `gpt-5.6-sol`, `xhigh` reasoning; requires the `codex` CLI >= 0.144.1.
 - **Domain guides**: `.claude/skills/native-trigger/` and `frontend/tutorial-system-guide.mdc`
 - **Brand/UI guidelines**: `frontend/brand-guidelines.md`
 - **CLI commands**: when adding/modifying/removing a command, subcommand, option, or description in `cli/src/commands/`, run `python system_prompts/generate.py` to refresh `system_prompts/auto-generated/` and `cli/src/guidance/skills.gen.ts`. The CLI docs the agents use to operate `wmill` are derived from the source — stale generated files give agents the wrong flags.
@@ -24,6 +24,8 @@ Open-source platform for internal tools, workflows, API integrations, background
 ## Dev Environment

 - **Backend**: `cargo run` from `backend/` (API at http://localhost:8000)
+- **DuckDB local jobs**: before running DuckDB scripts locally, build the FFI shared library with `cd backend/windmill-duckdb-ffi-internal && ./build_dev.sh`. Re-run it after clean builds or when `backend/target/debug/libwindmill_duckdb_ffi_internal.*` is missing. The bundled DuckDB compile (~2min) is cached in a per-user dir shared across worktrees, so a fresh worktree reuses it and the build is near-instant.
+- **Data pipelines (DuckLake) from source**: a plain `cargo run` (even `--features quickjs`) advertises a `duckdb` worker tag but **cannot** execute DuckDB scripts and has **no** working S3 proxy (DuckLake writes 404). Build CE DuckLake with `cargo run --features quickjs,duckdb,parquet,private` (add `,python` for Python scripts, `,enterprise,license` for EE) **and** build the FFI (bullet above). See `backend/CLAUDE.md` → "Running data pipelines (DuckLake) from source" for the exact feature sets and the two feature-gate gotchas.
 - **Frontend**: `REMOTE=http://localhost:8000 npm run dev` from `frontend/` (port 3000+)
 - **DB**: `psql postgres://postgres:changeme@localhost:5432/windmill`
 - **Login**: `admin@windmill.dev` / `changeme`
@@ -109,5 +111,6 @@ $NAV --root backend callees "X"                           # what does X call?
 - Search for existing code to reuse before writing new code
 - Follow established patterns in the codebase
 - Keep changes focused — don't refactor beyond what's asked
- **Comments record constraints, not narration.** Write a comment only for what the code can't show: why a non-obvious approach is required, what breaks if it's "simplified" away. State each invariant once, at the place where someone would break it, in ≤4 lines. Don't describe what the next line does, don't repeat the same rationale at multiple sites, and don't address the PR reviewer (justifying a change belongs in the PR description, not the code). Describe the code as it is, never its drafting history: "we no longer do X", "unchanged behavior", "instead of the previous approach" are meaningless to a reader who never saw the earlier iteration — before finishing, reread your comments as if the current state is the only state that ever existed.
+- **Ship only the tests the PR needs.** A committed test must pin behavior a future change could plausibly break, and be the smallest setup that exercises the new logic. While developing, write as many exhaustive tests and do as much manual testing as you need to convince yourself the change works — then remove that scaffolding before marking the PR ready, keeping only the essential regression guard(s). A test that merely re-exercises pre-existing behavior, or needs elaborate fixtures to assert something trivial, is scaffolding: delete it. If nothing meaningful is left to guard, ship no test rather than a ceremonial one.
+- **Comments record constraints, not narration.** Write a comment only for what the code can't show: why a non-obvious approach is required, what breaks if it's "simplified" away. State each invariant once, at the place where someone would break it, in ≤4 lines. Don't describe what the next line does, don't repeat the same rationale at multiple sites, and don't address the PR reviewer (justifying a change belongs in the PR description, not the code). Reference nothing ephemeral — no numbered steps from your dev flow, no "the poller / the test does X" scaffolding, no transient state that won't exist for the next reader; keep only the essential, durable rationale. Describe the code as it is, never its drafting history: "we no longer do X", "unchanged behavior", "instead of the previous approach" are meaningless to a reader who never saw the earlier iteration — before finishing, reread your comments as if the current state is the only state that ever existed.
 - **Never attribute work to a specific customer, account, or "requested by a customer" in repo-tracked content** (PR descriptions, commit messages, code comments, docs). Describe changes by their technical motivation instead.
@@ -1,5 +1,694 @@
 # Changelog

+## [1.766.1](https://github.com/windmill-labs/windmill/compare/v1.766.0...v1.766.1) (2026-07-22)
+
+
+### Bug Fixes
+
+* **jobs:** enforce self_approval_disabled on the UI resume path ([#10262](https://github.com/windmill-labs/windmill/issues/10262)) ([2d24b3a](https://github.com/windmill-labs/windmill/commit/2d24b3ac49b65469072bca929730d2c9ecd54b8f))
+
+## [1.766.0](https://github.com/windmill-labs/windmill/compare/v1.765.0...v1.766.0) (2026-07-22)
+
+
+### Features
+
+* **ai-sessions:** show item preview cards for tools ([#10254](https://github.com/windmill-labs/windmill/issues/10254)) ([703744f](https://github.com/windmill-labs/windmill/commit/703744fb8b76f70384cbc3341d5e1b3dbf30455e))
+* make content search a full CE feature ([#10252](https://github.com/windmill-labs/windmill/issues/10252)) ([d2c5d6f](https://github.com/windmill-labs/windmill/commit/d2c5d6f4b4ede9449407ed3831713d3e3d1d6972))
+* **sessions:** ship AI sessions as beta with legacy-chat opt-out ([#10242](https://github.com/windmill-labs/windmill/issues/10242)) ([0508cdd](https://github.com/windmill-labs/windmill/commit/0508cddf0a86b6a6a043cb31978ca68b76a95d02))
+
+
+### Bug Fixes
+
+* accept ssh/scheme-less git repo urls and $var: refs in app repo resolution ([#10246](https://github.com/windmill-labs/windmill/issues/10246)) ([b948efd](https://github.com/windmill-labs/windmill/commit/b948efd3c81aada103bd141eae03fbe626bf6ee5))
+* **copilot:** stop write_flow forcing rawscript code into nested JSON ([#10260](https://github.com/windmill-labs/windmill/issues/10260)) ([ecb1a92](https://github.com/windmill-labs/windmill/commit/ecb1a92070fb18048a6a1467270469e8462a3cbf))
+* prevent u16 underflow in suspend count causing permanent flow deadlock ([#10256](https://github.com/windmill-labs/windmill/issues/10256)) ([68b1fcc](https://github.com/windmill-labs/windmill/commit/68b1fcc5cdd3d217ebb1b9e09c0d52334faf0a63))
+* **prompts:** prefer Bun over Deno for TypeScript runtime selection ([#10253](https://github.com/windmill-labs/windmill/issues/10253)) ([380cf75](https://github.com/windmill-labs/windmill/commit/380cf752ca8fc57eb70f7c8efd57290f7902a030))
+* **tutorials:** repair broken frontend tutorials after UI redesigns ([#10255](https://github.com/windmill-labs/windmill/issues/10255)) ([5685981](https://github.com/windmill-labs/windmill/commit/5685981c9902ff37c3bd271dc5c5ede948a0787a))
+
+## [1.765.0](https://github.com/windmill-labs/windmill/compare/v1.764.0...v1.765.0) (2026-07-21)
+
+
+### Features
+
+* **ai:** open the Compare & Deploy page from chat with item preselection ([#10232](https://github.com/windmill-labs/windmill/issues/10232)) ([572d69e](https://github.com/windmill-labs/windmill/commit/572d69e5ae8ae12207dca175fdb5ea378d852a45))
+* **ai:** session chat nits — empty sends, command picker polish, session-state prompt ([#10233](https://github.com/windmill-labs/windmill/issues/10233)) ([9bc1f62](https://github.com/windmill-labs/windmill/commit/9bc1f6212837155c964a7db7103aabbcf14cbce2))
+* attach text files to chat messages, read on demand via file tools ([#10215](https://github.com/windmill-labs/windmill/issues/10215)) ([d6cf1ef](https://github.com/windmill-labs/windmill/commit/d6cf1ef9872cd0db0f51e4881237bac383e45f61))
+* **git-sync:** enable per-item promotion mode on dev workspaces ([#10205](https://github.com/windmill-labs/windmill/issues/10205)) ([2ce21c9](https://github.com/windmill-labs/windmill/commit/2ce21c9ef86c608d73e12a7e278d2f328e06b6fa))
+* **pipelines:** catalog declared measures and dimensions ([#10190](https://github.com/windmill-labs/windmill/issues/10190)) ([fd51d40](https://github.com/windmill-labs/windmill/commit/fd51d40f1254a206de1f38e59cc3dadce8e13aa5))
+* **triggers:** add AMQP (RabbitMQ) trigger via lapin ([#10230](https://github.com/windmill-labs/windmill/issues/10230)) ([68debab](https://github.com/windmill-labs/windmill/commit/68debab877c6dc8ee3732e0c23d467db85fd4584))
+* unified read-only `diff` chat tool (drafts, fork vs parent, search) ([#10211](https://github.com/windmill-labs/windmill/issues/10211)) ([9739d5a](https://github.com/windmill-labs/windmill/commit/9739d5a2c2a21beded8848ea9049b2816196a898))
+
+
+### Bug Fixes
+
+* **apps:** let entitled viewers read pre-existing S3 files from deployed apps ([#10245](https://github.com/windmill-labs/windmill/issues/10245)) ([4a89824](https://github.com/windmill-labs/windmill/commit/4a898247a21ae918fa952a993fbe407ebc49e404))
+* **db:** repair s3 asset paths missing default-storage leading slash ([#10243](https://github.com/windmill-labs/windmill/issues/10243)) ([555c751](https://github.com/windmill-labs/windmill/commit/555c751016fea4087be09fa2520af331cc872bfc))
+* **frontend:** curl fallback for +Variable/+Resource in bash sandbox mode ([#10235](https://github.com/windmill-labs/windmill/issues/10235)) ([28966bd](https://github.com/windmill-labs/windmill/commit/28966bdbf190ac461aa5259e04d6c15cf699fd94))
+* **frontend:** limit compare & deploy rows to the active direction ([#10234](https://github.com/windmill-labs/windmill/issues/10234)) ([7ac27c1](https://github.com/windmill-labs/windmill/commit/7ac27c1ef240729fc38d5fa506b96bf60be74e25))
+* **frontend:** prevent browser back-swipe navigation over monaco editors ([#10229](https://github.com/windmill-labs/windmill/issues/10229)) ([b0bf256](https://github.com/windmill-labs/windmill/commit/b0bf25683ba3aac3e94bbe9d6bbd6e22545e734e))
+* **parsers:** keep s3 asset path suffix verbatim to preserve storage distinction ([#10241](https://github.com/windmill-labs/windmill/issues/10241)) ([7fb8a2e](https://github.com/windmill-labs/windmill/commit/7fb8a2e390cef3cd01a33c89ee55d3f53cb1e20f))
+* **parser:** spurious pg arg inferred from placeholders in comments ([#10226](https://github.com/windmill-labs/windmill/issues/10226)) ([d24e176](https://github.com/windmill-labs/windmill/commit/d24e1768163c10fbbc144bd1126457e42d282e37))
+* **pg:** actionable error when s3object input exceeds jsonb 256MB cap ([#10228](https://github.com/windmill-labs/windmill/issues/10228)) ([6e42633](https://github.com/windmill-labs/windmill/commit/6e4263364325e4d4616697c6da601bbc887ff43c))
+* **postgres-triggers:** enforce resource-path scopes on ancillary routes ([#10222](https://github.com/windmill-labs/windmill/issues/10222)) ([39058c0](https://github.com/windmill-labs/windmill/commit/39058c0a01c35cf553daae3af37c6d326a7c8763))
+* return to parent workspace when a fork is deleted remotely ([#9898](https://github.com/windmill-labs/windmill/issues/9898)) ([32994df](https://github.com/windmill-labs/windmill/commit/32994df427e6df6747caf1ba6371512d5a5be5c2))
+* **s3:** support instance-policy credentials in object storage tests ([#10238](https://github.com/windmill-labs/windmill/issues/10238)) ([ec63244](https://github.com/windmill-labs/windmill/commit/ec6324409d6c0e9ddcda5d110ac570ca79fe9e38))
+* **triggers:** apply scope-path filtering to list and fix update scope check ([#10220](https://github.com/windmill-labs/windmill/issues/10220)) ([adc555d](https://github.com/windmill-labs/windmill/commit/adc555d1722cb59aa5b88503b3105357d76f5b71))
+* **worker:** mount /dev/shm as tmpfs in the Docker v2 nsjail sandbox ([#10240](https://github.com/windmill-labs/windmill/issues/10240)) ([2caee41](https://github.com/windmill-labs/windmill/commit/2caee41fdfe1e69010a1f4544d45aaa5db49f590))
+
+## [1.764.0](https://github.com/windmill-labs/windmill/compare/v1.763.0...v1.764.0) (2026-07-20)
+
+
+### Features
+
+* **ai:** add npm package search tool to global chat ([#10204](https://github.com/windmill-labs/windmill/issues/10204)) ([0593ff7](https://github.com/windmill-labs/windmill/commit/0593ff7d7d598af0b9f52d08bb22147e3a83e6e9))
+* **ai:** expose get_db_schema tool in global chat ([#10207](https://github.com/windmill-labs/windmill/issues/10207)) ([f4308cf](https://github.com/windmill-labs/windmill/commit/f4308cf033b75e9a69c1e21041ff97f3bfa1787c))
+* **ai:** extract prompt cache token usage from OpenAI and Azure providers ([#10214](https://github.com/windmill-labs/windmill/issues/10214)) ([2b58df5](https://github.com/windmill-labs/windmill/commit/2b58df57fc5bae08155ffd7198643ed739e6a463))
+* **ai:** live web-search source list on chat tool cards ([#10210](https://github.com/windmill-labs/windmill/issues/10210)) ([542a484](https://github.com/windmill-labs/windmill/commit/542a4842a317a6607de973b4567bdb27dc088664))
+* cap queued jobs per concurrency key on cloud ([#10197](https://github.com/windmill-labs/windmill/issues/10197)) ([71f2d47](https://github.com/windmill-labs/windmill/commit/71f2d47cb4ff331cbcc7843bf0704ddb63cb77db))
+* detect server-handled git-sync so CLI picks git push vs wmill sync push ([#10201](https://github.com/windmill-labs/windmill/issues/10201)) ([b070f56](https://github.com/windmill-labs/windmill/commit/b070f56c5e6109c219a53cbb9628e28be4397fb7))
+* expose windmill api endpoint catalog to global ai chat ([#10199](https://github.com/windmill-labs/windmill/issues/10199)) ([83a354f](https://github.com/windmill-labs/windmill/commit/83a354f831cbc1821f2316e6d4e1c7f7a2cd7cdb))
+* **jobs:** cap total queued jobs per workspace on cloud ([#10218](https://github.com/windmill-labs/windmill/issues/10218)) ([ddec2ab](https://github.com/windmill-labs/windmill/commit/ddec2abbb3f56fc44d4c75fde60f00fb1c6373bb))
+* **sessions:** live DOM access for the raw-app preview in AI sessions ([#10129](https://github.com/windmill-labs/windmill/issues/10129)) ([b448af1](https://github.com/windmill-labs/windmill/commit/b448af1da7f6a351973724721e52682259212292))
+* **telemetry:** generic feature-usage telemetry with AI session metrics ([#10200](https://github.com/windmill-labs/windmill/issues/10200)) ([11fda89](https://github.com/windmill-labs/windmill/commit/11fda89b520cc8d1ce30cdba36bd10355cf025cc))
+
+
+### Bug Fixes
+
+* **db:** grant schema usage and re-run windmill role grants ([#10212](https://github.com/windmill-labs/windmill/issues/10212)) ([38ceae1](https://github.com/windmill-labs/windmill/commit/38ceae1a098ae165c75f43aa9049b5afafdd3bac))
+* **forks:** show workspace settings link in sidebar for fork creators ([#10216](https://github.com/windmill-labs/windmill/issues/10216)) ([11bb37d](https://github.com/windmill-labs/windmill/commit/11bb37d7ca7fbd50cc66905f41851226a48e57fb))
+* **git-sync:** avoid percent-encoded colon in git-sync hub script path ([#10213](https://github.com/windmill-labs/windmill/issues/10213)) ([87be041](https://github.com/windmill-labs/windmill/commit/87be041c091dd7fdd8c6be907a9b2894779f9ac0))
+* **inputs:** add ownership check to saved-input args read endpoint ([#10194](https://github.com/windmill-labs/windmill/issues/10194)) ([f32d770](https://github.com/windmill-labs/windmill/commit/f32d7702bcb592a70f5cf7e01f9e830047645240))
+* **schedules:** apply scope-path filtering to schedule list endpoints ([#10192](https://github.com/windmill-labs/windmill/issues/10192)) ([70359e3](https://github.com/windmill-labs/windmill/commit/70359e3a76d2e1fa5994acce4909753410c80664))
+* **security:** enforce token scope filtering on folder list endpoints ([#10193](https://github.com/windmill-labs/windmill/issues/10193)) ([b5e69ff](https://github.com/windmill-labs/windmill/commit/b5e69ffba6afd49d147e9b5e88e2d7020c09142f))
+* **sessions:** session bar badge readouts, job persistence, refresh bounce ([#10217](https://github.com/windmill-labs/windmill/issues/10217)) ([c8870d3](https://github.com/windmill-labs/windmill/commit/c8870d36aed2283bfa14419cd0bb346083f27b9c))
+* steer ai chat away from draft-blind api catalog reads and runs ([#10202](https://github.com/windmill-labs/windmill/issues/10202)) ([0e04bc6](https://github.com/windmill-labs/windmill/commit/0e04bc6991486b2b2b05b13feed5b083e403ac80))
+
+## [1.763.0](https://github.com/windmill-labs/windmill/compare/v1.762.2...v1.763.0) (2026-07-19)
+
+
+### Features
+
+* **ai:** gate data pipelines in sessions behind a dev flag ([#10178](https://github.com/windmill-labs/windmill/issues/10178)) ([42da20a](https://github.com/windmill-labs/windmill/commit/42da20ae9743eab2f86b8f9c261ee567f8bbb0c5))
+
+
+### Bug Fixes
+
+* **cli:** stop emitting has_on_behalf_of/has_permissioned_as: false on pull ([#10188](https://github.com/windmill-labs/windmill/issues/10188)) ([2d77e74](https://github.com/windmill-labs/windmill/commit/2d77e742077b71be192a04ed9e85d4a024dc7f91))
+
+## [1.762.2](https://github.com/windmill-labs/windmill/compare/v1.762.1...v1.762.2) (2026-07-18)
+
+
+### Bug Fixes
+
+* **scripts:** populate auto_kind from draft JSON for draft-only scripts ([#10183](https://github.com/windmill-labs/windmill/issues/10183)) ([cba5f0d](https://github.com/windmill-labs/windmill/commit/cba5f0d8a8c2261a190d0f10396a6ad6aed89366))
+
+## [1.762.1](https://github.com/windmill-labs/windmill/compare/v1.762.0...v1.762.1) (2026-07-17)
+
+
+### Bug Fixes
+
+* **schedules:** stop disabling schedules on transient push errors ([#10179](https://github.com/windmill-labs/windmill/issues/10179)) ([c82056c](https://github.com/windmill-labs/windmill/commit/c82056cfde93ab0e26cf82aa4f7650c060565542))
+* **worker:** gate ansible socket-dir name check to unix ([#10180](https://github.com/windmill-labs/windmill/issues/10180)) ([203f6c6](https://github.com/windmill-labs/windmill/commit/203f6c69dd24fc055a4bf604315aa757f35d0565))
+
+## [1.762.0](https://github.com/windmill-labs/windmill/compare/v1.761.0...v1.762.0) (2026-07-17)
+
+
+### Features
+
+* **ai-chat:** image attachments and agent raw-app screenshots ([#10130](https://github.com/windmill-labs/windmill/issues/10130)) ([7a139ab](https://github.com/windmill-labs/windmill/commit/7a139ab23e034321554d0c85540d54251e61cc72))
+* **forks:** let a fork's creator manage developers on it without being an admin ([#10166](https://github.com/windmill-labs/windmill/issues/10166)) ([97f4477](https://github.com/windmill-labs/windmill/commit/97f44770698eb3dd9e296012360b7d9a4492e28d))
+* **otel-tracing-proxy:** trust internal endpoints with untrusted CAs ([#10139](https://github.com/windmill-labs/windmill/issues/10139)) ([396fb1c](https://github.com/windmill-labs/windmill/commit/396fb1c4752b2b93ef0ec9d0c9c6be05aa403eef))
+* **worker-tags:** add `*` fork marker to workspace-scoped custom tags ([#10177](https://github.com/windmill-labs/windmill/issues/10177)) ([2ff5a91](https://github.com/windmill-labs/windmill/commit/2ff5a918d54074bc340b6e92222e9f5bb4369ba0))
+
+
+### Bug Fixes
+
+* **ai:** unbreak session chat compaction for Anthropic models ([#10171](https://github.com/windmill-labs/windmill/issues/10171)) ([a9fc9f7](https://github.com/windmill-labs/windmill/commit/a9fc9f74b2c5b280cc087f0bfbac50112081c5e8))
+* **ansible:** keep persistent-connection socket path under the AF_UNIX limit ([#10167](https://github.com/windmill-labs/windmill/issues/10167)) ([be57dd9](https://github.com/windmill-labs/windmill/commit/be57dd91e40551ff0d6fbbeccd97e2b98589c5b9))
+* **extra:** make the extra container runnable as a non-root UID ([#10173](https://github.com/windmill-labs/windmill/issues/10173)) ([dcb9e40](https://github.com/windmill-labs/windmill/commit/dcb9e40ea237c3744611ac6e83ff5ecc6528e697))
+* **flows:** make updateFlow body path optional so AI can update flows ([#10176](https://github.com/windmill-labs/windmill/issues/10176)) ([7d2c5ce](https://github.com/windmill-labs/windmill/commit/7d2c5ceb0fb8609aa58a1e28e61ae8ce86e48208))
+* **frontend:** scope session pipeline trigger editors to the session workspace ([#10032](https://github.com/windmill-labs/windmill/issues/10032)) ([8828341](https://github.com/windmill-labs/windmill/commit/8828341a2b0912cbf44d6c6d81b4f4e30c5fe631))
+* **mcp:** apply token scopes consistently across mcp endpoint tools ([#10162](https://github.com/windmill-labs/windmill/issues/10162)) ([ae3d9ce](https://github.com/windmill-labs/windmill/commit/ae3d9ce2c031beeafc4e8156e0c85f163647af0f))
+* parse all names in grouped go param declarations ([#10165](https://github.com/windmill-labs/windmill/issues/10165)) ([d0aa7dc](https://github.com/windmill-labs/windmill/commit/d0aa7dca13aedfec90b7e023954cc824dafa3d34))
+* **schedules:** re-arm enabled schedules left with no queued occurrence ([#10174](https://github.com/windmill-labs/windmill/issues/10174)) ([9762089](https://github.com/windmill-labs/windmill/commit/9762089fcbdb33dbf453d22342661e5857b2ad7e))
+
+
+### Performance Improvements
+
+* lazy-load session preview editor views for snappy AI sessions switch ([#10172](https://github.com/windmill-labs/windmill/issues/10172)) ([1edee8a](https://github.com/windmill-labs/windmill/commit/1edee8aa34503d2892c1ef7b1deabf99e4b528b4))
+
+## [1.761.0](https://github.com/windmill-labs/windmill/compare/v1.760.1...v1.761.0) (2026-07-16)
+
+
+### Features
+
+* **ai-sessions:** CRUD markdown artifacts in sessions ([#10046](https://github.com/windmill-labs/windmill/issues/10046)) ([0ea5705](https://github.com/windmill-labs/windmill/commit/0ea570570e65abf64b3b171e592a8dd3eea0b105))
+* **alerts:** include disk total and top consumers in low-disk alert ([#10144](https://github.com/windmill-labs/windmill/issues/10144)) ([4e0fd4d](https://github.com/windmill-labs/windmill/commit/4e0fd4db5589f91426ca025184da7b977d5a16d6))
+* automatic git-to-windmill sync (polling, webhooks, in-app PRs + checks) ([#9552](https://github.com/windmill-labs/windmill/issues/9552)) ([51d8db6](https://github.com/windmill-labs/windmill/commit/51d8db6602bc192879e50b54a5fe6b6b8beb63f4))
+* display openai reasoning summaries in ai chat ([#10147](https://github.com/windmill-labs/windmill/issues/10147)) ([4ee1d32](https://github.com/windmill-labs/windmill/commit/4ee1d32101eb527ef707c3c4af456892f0a059ab))
+* **forks:** add "Hide unchanged drafts" toggle to fork deploy-draft tab ([#10022](https://github.com/windmill-labs/windmill/issues/10022)) ([4fc3f30](https://github.com/windmill-labs/windmill/commit/4fc3f304c632eda15876c28cb6f473a799a41f48))
+* **frontend:** flatten workspace pickers, whole-tab picker trigger ([#10145](https://github.com/windmill-labs/windmill/issues/10145)) ([7fda6a0](https://github.com/windmill-labs/windmill/commit/7fda6a05345d14f62dc2124d682650d50eecbe98))
+
+
+### Bug Fixes
+
+* **ai:** show the question in askUserQuestion tool-call labels ([#10153](https://github.com/windmill-labs/windmill/issues/10153)) ([fa03984](https://github.com/windmill-labs/windmill/commit/fa03984a14bb71f0ea2a27bfa23339a9d0d8e86b))
+* **alerts:** identify server replica in low-disk alert + per-host dedup tag ([#10143](https://github.com/windmill-labs/windmill/issues/10143)) ([3bd9f05](https://github.com/windmill-labs/windmill/commit/3bd9f0593866740935f79c8b2c1cb92829059105))
+* **frontend:** sanitize job result markup, gate it on unsandboxed public apps ([#10127](https://github.com/windmill-labs/windmill/issues/10127)) ([7b813d1](https://github.com/windmill-labs/windmill/commit/7b813d1f74baef200d9f1387e02925083b79b265))
+* **frontend:** show friendly draft path for draft-only items in pickers ([#10136](https://github.com/windmill-labs/windmill/issues/10136)) ([568dbbe](https://github.com/windmill-labs/windmill/commit/568dbbee852d59986581b0d16f5ac3e06d6bad6a))
+* **frontend:** surface real tool call errors in AI chat ([#10146](https://github.com/windmill-labs/windmill/issues/10146)) ([0694b84](https://github.com/windmill-labs/windmill/commit/0694b84da7bbb9404fd386e3ce109a574a0e9c93))
+* heartbeat job ping during s3object materialization in SQL executors ([#10152](https://github.com/windmill-labs/windmill/issues/10152)) ([7d5009e](https://github.com/windmill-labs/windmill/commit/7d5009e3928bb200150eb998ed0b3016c3395987))
+* **mcp:** push granular scope patterns into SQL so scoped scripts/flows aren't truncated ([#10140](https://github.com/windmill-labs/windmill/issues/10140)) ([91d6606](https://github.com/windmill-labs/windmill/commit/91d6606868d0e9c5f78ab48a592ef95ffaeeca61))
+* **migrations:** grant zombie_job_counter to windmill roles ([#10159](https://github.com/windmill-labs/windmill/issues/10159)) ([0e547ad](https://github.com/windmill-labs/windmill/commit/0e547adf23b615f0caccc88eb9101bb0604bb51b))
+* **raw-apps:** full code ui builder improvements ([c55ac53](https://github.com/windmill-labs/windmill/commit/c55ac5326fbb4532f22b7250a58b69396a4c868a))
+* **raw-apps:** prevent and surface the silent blank screen from an unmounted #root ([#10150](https://github.com/windmill-labs/windmill/issues/10150)) ([24750e6](https://github.com/windmill-labs/windmill/commit/24750e6ef1a1a975e0063df6809774f913b5d01b))
+* **self-host:** unbreak self-hosted Caddy after the caddy-l4 syntax change ([#10156](https://github.com/windmill-labs/windmill/issues/10156)) ([2f6c35b](https://github.com/windmill-labs/windmill/commit/2f6c35b15bf70f17eb8ff08281e0ecf8313e7b35))
+
+## [1.760.1](https://github.com/windmill-labs/windmill/compare/v1.760.0...v1.760.1) (2026-07-15)
+
+
+### Bug Fixes
+
+* **apps:** honor presigned S3 signature on app display/preview routes ([#10141](https://github.com/windmill-labs/windmill/issues/10141)) ([8c725d9](https://github.com/windmill-labs/windmill/commit/8c725d9e44e38bf3b35f41b847cdf8745e07942e))
+
+## [1.760.0](https://github.com/windmill-labs/windmill/compare/v1.759.0...v1.760.0) (2026-07-15)
+
+
+### Features
+
+* **nsjail:** make python/ansible rlimit_as configurable per worker (GIT-921) ([#10138](https://github.com/windmill-labs/windmill/issues/10138)) ([1787201](https://github.com/windmill-labs/windmill/commit/17872018cc037e699b1f6e1589d0ad05e0883cca))
+
+
+### Bug Fixes
+
+* **ai:** disable redirects on worker AI provider client (GHSA-5q4v) ([#10122](https://github.com/windmill-labs/windmill/issues/10122)) ([27ead8d](https://github.com/windmill-labs/windmill/commit/27ead8d0848cceacaf0c49fed0e8896472b851e3))
+* **ai:** stop sending the AI agent system prompt twice for OpenAI ([#10126](https://github.com/windmill-labs/windmill/issues/10126)) ([8bfe5c9](https://github.com/windmill-labs/windmill/commit/8bfe5c93404ba3f137394f16d0571a06d891dc3b))
+* **apps:** invalidate cached app policy on change or deletion (GHSA-r5v4-cxh9-7qhq) ([#10121](https://github.com/windmill-labs/windmill/issues/10121)) ([f7eb5c4](https://github.com/windmill-labs/windmill/commit/f7eb5c460d78792c24297e20cb062638522a4f68))
+* **bash:** normalize CRLF line endings before running scripts ([#10131](https://github.com/windmill-labs/windmill/issues/10131)) ([6407d9f](https://github.com/windmill-labs/windmill/commit/6407d9ff5ce51e71ff8b8fc503d89a2bdc2e1761))
+* **cli-image:** patch fixable CRITICAL CVEs in windmill-cli image (GIT-922) ([#10135](https://github.com/windmill-labs/windmill/issues/10135)) ([5626768](https://github.com/windmill-labs/windmill/commit/56267684718944ef4d4ecd3b610bad81132e1990))
+* **frontend:** graceful small-screen timeframe picker on the runs page ([#10073](https://github.com/windmill-labs/windmill/issues/10073)) ([af177ce](https://github.com/windmill-labs/windmill/commit/af177cefe07e6037e33cc90757088a98fb63a49f))
+* **frontend:** keep session-exit URL clean by syncing new_draft strip with the router ([#10101](https://github.com/windmill-labs/windmill/issues/10101)) ([9705d60](https://github.com/windmill-labs/windmill/commit/9705d602848966f850613233d6e23b73a753c259))
+* **frontend:** only carry custom-tag overrides on 'Run again' ([#10137](https://github.com/windmill-labs/windmill/issues/10137)) ([bd3adc9](https://github.com/windmill-labs/windmill/commit/bd3adc9781d8e77928c9feb03d0e05b63a1aaf7c))
+* **frontend:** treat a displaced draft save as superseded, not failed ([#10094](https://github.com/windmill-labs/windmill/issues/10094)) ([2fe999f](https://github.com/windmill-labs/windmill/commit/2fe999f66cd15acd81850f970ada31e9892abff2))
+* reject git URL fragment/query SSRF bypass (GHSA-p5cj-8cfh-mjv6) ([#10120](https://github.com/windmill-labs/windmill/issues/10120)) ([73c8d7f](https://github.com/windmill-labs/windmill/commit/73c8d7f08ad55cd2323d1bb9438f00a8ac30e046))
+* **security:** enforce variables:write scope on resource-delete var cascade (GHSA-xmr2-98m6-cjf7) ([#10123](https://github.com/windmill-labs/windmill/issues/10123)) ([188647a](https://github.com/windmill-labs/windmill/commit/188647a942a0cb496d63c71bf3e0f7a3136ec217))
+
+## [1.759.0](https://github.com/windmill-labs/windmill/compare/v1.758.0...v1.759.0) (2026-07-15)
+
+
+### Features
+
+* **nativets:** add Web Crypto support via deno_crypto ([#10109](https://github.com/windmill-labs/windmill/issues/10109)) ([ba23254](https://github.com/windmill-labs/windmill/commit/ba232544e7608731560defc0ed103c3e746e46ea))
+* **nativets:** expose the standard web-platform globals deno_web provides ([#10112](https://github.com/windmill-labs/windmill/issues/10112)) ([6d1e12d](https://github.com/windmill-labs/windmill/commit/6d1e12d5e95a6e3fb74c142e33a141b5ce0856a9))
+
+
+### Bug Fixes
+
+* **jseval:** raise QuickJS eval memory cap to 128MB with clear OOM error ([#10116](https://github.com/windmill-labs/windmill/issues/10116)) ([95d9ff0](https://github.com/windmill-labs/windmill/commit/95d9ff02ee8a92162c06857bac7102c92c708c51))
+* **mcp:** advertise flow input variables in MCP tools ([#10117](https://github.com/windmill-labs/windmill/issues/10117)) ([a6191e2](https://github.com/windmill-labs/windmill/commit/a6191e2a855e03d0b03847067787a9de26e9d54a))
+* **mcp:** let MCP tokens call preview run tools (jobs:run scope) — Fixes GIT-920 ([#10107](https://github.com/windmill-labs/windmill/issues/10107)) ([4917f79](https://github.com/windmill-labs/windmill/commit/4917f79935acd4ecf9fb5a21d3131dec9ae9da44))
+* **nativets:** apply parameter defaults for missing args instead of null ([#10111](https://github.com/windmill-labs/windmill/issues/10111)) ([ba7f9c0](https://github.com/windmill-labs/windmill/commit/ba7f9c065f78641487c2765370be98227dcf1c6a))
+* **self-host:** resolve caddy-l4 "unrecognized global option: layer4" error ([#10106](https://github.com/windmill-labs/windmill/issues/10106)) ([770ac2b](https://github.com/windmill-labs/windmill/commit/770ac2be9ed3ff8f647aa3948e1f42d62c97865b))
+* **tree-view:** align file indentation with sibling folders ([#10115](https://github.com/windmill-labs/windmill/issues/10115)) ([88030d0](https://github.com/windmill-labs/windmill/commit/88030d0f557a834f2dae80e2a31033261d8b1d1d))
+
+
+### Performance Improvements
+
+* **rls:** wrap session GUC reads in RLS policies for per-statement InitPlan (GIT-919) ([#10110](https://github.com/windmill-labs/windmill/issues/10110)) ([e9fd4e7](https://github.com/windmill-labs/windmill/commit/e9fd4e7554f624a7e6c3923b9c6b604bee9ebd1f))
+
+## [1.758.0](https://github.com/windmill-labs/windmill/compare/v1.757.0...v1.758.0) (2026-07-14)
+
+
+### Features
+
+* **ai-agent:** give tools a real description instead of the tool name ([#10083](https://github.com/windmill-labs/windmill/issues/10083)) ([7ebfad3](https://github.com/windmill-labs/windmill/commit/7ebfad382a2649a325444fcb745aca415871fe77))
+* **ai-chat:** port flow-group and sticky-note instructions to global chat ([#10090](https://github.com/windmill-labs/windmill/issues/10090)) ([32f32d9](https://github.com/windmill-labs/windmill/commit/32f32d9a29bb214d6aae58c501b8892ceb1c6453))
+* **cli:** add --tag override to script and flow run/preview ([#10079](https://github.com/windmill-labs/windmill/issues/10079)) ([98e6cca](https://github.com/windmill-labs/windmill/commit/98e6cca75d4dbc7417c7dd64289e0e8e56b84b0e))
+* **sessions:** support many pending sessions persisted in IndexedDB ([#10076](https://github.com/windmill-labs/windmill/issues/10076)) ([bfcec7e](https://github.com/windmill-labs/windmill/commit/bfcec7e8ac71e86ab16d7db789556b5fc7cfd3c7))
+
+
+### Bug Fixes
+
+* **ai-chat:** size AI-created flow notes to fit their text ([#10091](https://github.com/windmill-labs/windmill/issues/10091)) ([af3e3fe](https://github.com/windmill-labs/windmill/commit/af3e3fe6674de21a03b5c157a649452320c1ed0e))
+* **apps:** allow setting sandbox isolation and public access before first deploy ([#10085](https://github.com/windmill-labs/windmill/issues/10085)) ([cfc3f29](https://github.com/windmill-labs/windmill/commit/cfc3f292ad2fdc6067c558e42ef0754eca9469a9))
+* **apps:** load themes when selecting the Resources → Theme tab ([#10086](https://github.com/windmill-labs/windmill/issues/10086)) ([f2869d8](https://github.com/windmill-labs/windmill/commit/f2869d8c1a6f84837168d59724a496b39080bcce))
+* **frontend:** stop spurious raw-app reload that 404s on "Start without AI" ([#10099](https://github.com/windmill-labs/windmill/issues/10099)) ([2c702ef](https://github.com/windmill-labs/windmill/commit/2c702efec026bebdd9c1b8cdca4808e394b2fd09))
+* **mcp:** align script auto_kind filter with scripts list API ([#10098](https://github.com/windmill-labs/windmill/issues/10098)) ([4edffeb](https://github.com/windmill-labs/windmill/commit/4edffeb84b5d691884dc3c274dfd8aa4e9441295))
+* **sessions:** pending-draft debounce follow-ups (delete-cancel, keystroke de-transient, teardown count) ([#10087](https://github.com/windmill-labs/windmill/issues/10087)) ([f3cd5d9](https://github.com/windmill-labs/windmill/commit/f3cd5d9f7f370ba0aa27c452e8434454a10e07fe))
+* **sessions:** reopen script test panel when preview goes full screen ([#10082](https://github.com/windmill-labs/windmill/issues/10082)) ([eff9076](https://github.com/windmill-labs/windmill/commit/eff9076e9127eaa9b6a47897d5664932c720990b))
+
+
+### Performance Improvements
+
+* **runs:** index-bound batch re-run selection with a lossless completed_at bound (WIN-2168) ([#10074](https://github.com/windmill-labs/windmill/issues/10074)) ([15391f6](https://github.com/windmill-labs/windmill/commit/15391f6399eef5b85dac116b3ae04e71f70263a1))
+
+## [1.757.0](https://github.com/windmill-labs/windmill/compare/v1.756.1...v1.757.0) (2026-07-14)
+
+
+### Features
+
+* **saml:** add ALLOW_PRIVATE_SAML_METADATA_URLS SSRF bypass ([#10077](https://github.com/windmill-labs/windmill/issues/10077)) ([851e309](https://github.com/windmill-labs/windmill/commit/851e30914e3172dad2a3ccaa6951a3ffe1e67495))
+
+
+### Bug Fixes
+
+* **ai-agent:** don't mark repeated tool calls as failed in flow graph ([#10075](https://github.com/windmill-labs/windmill/issues/10075)) ([207ce86](https://github.com/windmill-labs/windmill/commit/207ce8649cf7026c62eea2b1b2f462c7df8c4e5a))
+
+## [1.756.1](https://github.com/windmill-labs/windmill/compare/v1.756.0...v1.756.1) (2026-07-14)
+
+
+### Bug Fixes
+
+* **apps:** cover script/flow component outputs in deployed-app S3 provenance gate ([#10070](https://github.com/windmill-labs/windmill/issues/10070)) ([710a13a](https://github.com/windmill-labs/windmill/commit/710a13a59d511614839f70e445db736e436550d7))
+
+## [1.756.0](https://github.com/windmill-labs/windmill/compare/v1.755.0...v1.756.0) (2026-07-12)
+
+
+### Features
+
+* **triggers:** serve binary HTTP-route responses via base64 transfer encoding ([#10058](https://github.com/windmill-labs/windmill/issues/10058)) ([29f4cd4](https://github.com/windmill-labs/windmill/commit/29f4cd4b6f58a29b83b84a7a9b8a439d20ade00e)), closes [#5986](https://github.com/windmill-labs/windmill/issues/5986)
+
+
+### Bug Fixes
+
+* replicate all secrets on fork when external backend is configured ([#10060](https://github.com/windmill-labs/windmill/issues/10060)) ([92b7f37](https://github.com/windmill-labs/windmill/commit/92b7f375a90de2f78565ca06a13c79ff04eda44d))
+* **sessions:** sync AI-session preview with workspace edits + stop phantom autosave (WIN-2160) ([#10061](https://github.com/windmill-labs/windmill/issues/10061)) ([5cde2d5](https://github.com/windmill-labs/windmill/commit/5cde2d5b6746be9f2d0be3a98ecdaf08777a6395))
+
+## [1.755.0](https://github.com/windmill-labs/windmill/compare/v1.754.0...v1.755.0) (2026-07-11)
+
+
+### Features
+
+* add per-workspace job-retention override ([#10050](https://github.com/windmill-labs/windmill/issues/10050)) ([ff774c4](https://github.com/windmill-labs/windmill/commit/ff774c46bff4bff1c532e512b163225aa7c41c11))
+* **apps:** authorize deployed-app S3 reads on-behalf of the author for logged-in viewers ([#10048](https://github.com/windmill-labs/windmill/issues/10048)) ([1e192f2](https://github.com/windmill-labs/windmill/commit/1e192f2d864b8a4671e900726972737406bc388a))
+* **mcp:** add multi-workspace MCP tokens via the gateway endpoint ([#10043](https://github.com/windmill-labs/windmill/issues/10043)) ([8343203](https://github.com/windmill-labs/windmill/commit/8343203ec2cea28a2ffd5b4ac636497e861fa3ce))
+
+
+### Bug Fixes
+
+* clearer errors on auto-draft save failure (WIN-2157) ([#10053](https://github.com/windmill-labs/windmill/issues/10053)) ([04eb7dd](https://github.com/windmill-labs/windmill/commit/04eb7ddd3906c28bec1711e276473a87c7b9500f))
+* **docker:** pin ansible tool interpreter to a persistent path ([#10054](https://github.com/windmill-labs/windmill/issues/10054)) ([6f49a1f](https://github.com/windmill-labs/windmill/commit/6f49a1f6a904442fcae9bb703f095b0a3ef61268))
+* enforce read authorization when signing S3 objects ([#10049](https://github.com/windmill-labs/windmill/issues/10049)) ([5844c32](https://github.com/windmill-labs/windmill/commit/5844c32ac5d08081b3de7f3d11b8b98eb1e1ad9a))
+* **frontend:** don't re-seed empty editor on stale ?new_draft after draft exists ([#10044](https://github.com/windmill-labs/windmill/issues/10044)) ([e668193](https://github.com/windmill-labs/windmill/commit/e668193a93b4a7df50459b31f2dc5f9a9b23d0fe))
+* **frontend:** keep draft autosave alive after AI-session round-trip ([#10052](https://github.com/windmill-labs/windmill/issues/10052)) ([7d02d9a](https://github.com/windmill-labs/windmill/commit/7d02d9a1e47760f287a71f6e09cd6fe45efb5635))
+* **frontend:** mint draft path for new SDK builder items so autosave attaches ([#10056](https://github.com/windmill-labs/windmill/issues/10056)) ([a89b896](https://github.com/windmill-labs/windmill/commit/a89b896ce5638f42f334055f9ffe6971b047aa84))
+* **frontend:** show nested restart button for subflows nested in containers ([#10042](https://github.com/windmill-labs/windmill/issues/10042)) ([3b07817](https://github.com/windmill-labs/windmill/commit/3b0781761b70667c5961bcb15d41c907716fa9e7))
+* **frontend:** show optimistic user message and fork-creation label before beforeSend ([#10037](https://github.com/windmill-labs/windmill/issues/10037)) ([1c88242](https://github.com/windmill-labs/windmill/commit/1c88242849a02b927f59e0a67c4b4707371b784f))
+* keep agent-worker server job-completed processor alive & self-healing ([#10033](https://github.com/windmill-labs/windmill/issues/10033)) ([ab38e14](https://github.com/windmill-labs/windmill/commit/ab38e1418e67be8bcc37391bb21d2f86d1ca3fc6))
+
+## [1.754.0](https://github.com/windmill-labs/windmill/compare/v1.753.0...v1.754.0) (2026-07-10)
+
+
+### Features
+
+* add multi-select mode to copilot askUserQuestion ([#10016](https://github.com/windmill-labs/windmill/issues/10016)) ([7569798](https://github.com/windmill-labs/windmill/commit/756979852c3245d06c5c73eb60e9a09fd59635c5))
+
+
+### Bug Fixes
+
+* accept bunnative language in AI chat flow step validation ([#10030](https://github.com/windmill-labs/windmill/issues/10030)) ([5a460db](https://github.com/windmill-labs/windmill/commit/5a460dbec6e2b81e01aa2c36cb504dde4ff6b24a))
+* **backend:** propagate script timeout when restarting perpetual scripts ([#10029](https://github.com/windmill-labs/windmill/issues/10029)) ([6c521e9](https://github.com/windmill-labs/windmill/commit/6c521e9d87724e43ebe3b7fce8b30a3671ef3d89))
+* **frontend:** name the draft in AI chat test-run confirmation ([#10024](https://github.com/windmill-labs/windmill/issues/10024)) ([9036ac7](https://github.com/windmill-labs/windmill/commit/9036ac789f358103b8d639a6b94708c452f52f90))
+* **frontend:** open new script/flow/app in AI session (not-found + friendly tab) ([#10028](https://github.com/windmill-labs/windmill/issues/10028)) ([0353569](https://github.com/windmill-labs/windmill/commit/03535691d607d8a9d1c1fa1c9c284fa3b6051d40))
+* **frontend:** persist forked "Copy of X" script drafts ([#10021](https://github.com/windmill-labs/windmill/issues/10021)) ([c537d45](https://github.com/windmill-labs/windmill/commit/c537d45e4982f30a44026d6644d5340e56f16ef9))
+* **frontend:** persist per-session preview panel resize width ([#10031](https://github.com/windmill-labs/windmill/issues/10031)) ([5387076](https://github.com/windmill-labs/windmill/commit/5387076c1c6fb35a99843aadd2055d3ac381d6cf))
+* **frontend:** scope raw-app, flow and script editors to the session workspace ([#10015](https://github.com/windmill-labs/windmill/issues/10015)) ([c000bbc](https://github.com/windmill-labs/windmill/commit/c000bbca283f5d61cff8a39458764b2b2dd2b58f))
+* resolve fork family/picker for superadmin visiting a non-member workspace ([#10023](https://github.com/windmill-labs/windmill/issues/10023)) ([368fd2d](https://github.com/windmill-labs/windmill/commit/368fd2d9e4b3ffb66e64934d9a622eea291cde5a))
+* scope AI-session flow/script editors to the session workspace ([#10025](https://github.com/windmill-labs/windmill/issues/10025)) ([c5060a1](https://github.com/windmill-labs/windmill/commit/c5060a1e9af5a704e90f92d325abf29626ecd28a))
+* **security:** drop --allow-run from Deno sandbox (GHSA-gj6h-vw66-mr8f) ([#10039](https://github.com/windmill-labs/windmill/issues/10039)) ([c029d6d](https://github.com/windmill-labs/windmill/commit/c029d6dcde44a3d16dee23a80afad920a0535b73))
+* **security:** remove git from Deno sandbox allow-run (GHSA-gj6h-vw66-mr8f) ([#10038](https://github.com/windmill-labs/windmill/issues/10038)) ([689b20a](https://github.com/windmill-labs/windmill/commit/689b20a4704a3dda8d9437b4793c0d16eb1f780f))
+* session preview tab labels, splitter hover, and diff-drawer sizing ([#10008](https://github.com/windmill-labs/windmill/issues/10008)) ([c139eed](https://github.com/windmill-labs/windmill/commit/c139eed631548113b843b466f5505bf6a01f17d3))
+* **sessions:** open test pane when enabling debug so the debug UI is visible ([#9998](https://github.com/windmill-labs/windmill/issues/9998)) ([d7a9b46](https://github.com/windmill-labs/windmill/commit/d7a9b46ab95108c7669b47b7c4be6d8c7964a9e6))
+* sync theme into session page preview iframes on toggle ([#10018](https://github.com/windmill-labs/windmill/issues/10018)) ([3704d00](https://github.com/windmill-labs/windmill/commit/3704d00956dea3b8e562a894d5330e052a753cb3))
+
+
+### Performance Improvements
+
+* index v2_job(parent_job) to speed up run child-job listing ([#10034](https://github.com/windmill-labs/windmill/issues/10034)) ([9feda57](https://github.com/windmill-labs/windmill/commit/9feda57c15bddc7ef481579b73636b88c2a143c5))
+* skip redundant retry-chain job query for successful top-level scripts ([#10035](https://github.com/windmill-labs/windmill/issues/10035)) ([15f9e9b](https://github.com/windmill-labs/windmill/commit/15f9e9b48fc326aa3d776191aabaed22f4c41e74))
+
+## [1.753.0](https://github.com/windmill-labs/windmill/compare/v1.752.0...v1.753.0) (2026-07-08)
+
+
+### Features
+
+* AI chat background jobs tray with detach, approval and preview ([#9982](https://github.com/windmill-labs/windmill/issues/9982)) ([286da00](https://github.com/windmill-labs/windmill/commit/286da005ef2faad1d193640f74f8e96999358707))
+* condensed top bar for session preview editors ([#10011](https://github.com/windmill-labs/windmill/issues/10011)) ([b847ca2](https://github.com/windmill-labs/windmill/commit/b847ca2bc7f06f494aa802d4350d6f032ba2bb58))
+* **db-health:** add connection sizing guidance ([#10014](https://github.com/windmill-labs/windmill/issues/10014)) ([f28ea9c](https://github.com/windmill-labs/windmill/commit/f28ea9cb991bbda32ed1b7a37a5f1b3552a589a8))
+* **sessions:** scoped preview refresh + multi-target live editors + pipeline preview ([#10006](https://github.com/windmill-labs/windmill/issues/10006)) ([32c398f](https://github.com/windmill-labs/windmill/commit/32c398f27de8cd5b1478ef60d247c13705b6b50f))
+* shared tab system, universal markdown code blocks, subtle scrollbars ([#10003](https://github.com/windmill-labs/windmill/issues/10003)) ([a00ee51](https://github.com/windmill-labs/windmill/commit/a00ee5196b2c013e9672ab029f5477079ac5da21))
+
+
+### Bug Fixes
+
+* bump bundled Go CLIs to patched versions to clear image CVEs ([#9996](https://github.com/windmill-labs/windmill/issues/9996)) ([d467161](https://github.com/windmill-labs/windmill/commit/d467161117444d7d9b18def627e90d9622512e02))
+* name the offending item when a fork fails on a NUL escape ([#10013](https://github.com/windmill-labs/windmill/issues/10013)) ([99d0047](https://github.com/windmill-labs/windmill/commit/99d00475156def6faad255c4e728923253f9169f))
+* preserve worker group tag override on 'Run again' ([#10004](https://github.com/windmill-labs/windmill/issues/10004)) ([c4cb2f3](https://github.com/windmill-labs/windmill/commit/c4cb2f373b6361f0f3ce6b1c8e32a4c010207760))
+* replicate external secret backend secrets when forking a workspace ([#10007](https://github.com/windmill-labs/windmill/issues/10007)) ([f65fe7b](https://github.com/windmill-labs/windmill/commit/f65fe7bf585d353f7d88746e947d68e2f351e516))
+* session preview editors and picker dropdown overflow ([#10010](https://github.com/windmill-labs/windmill/issues/10010)) ([fb12b23](https://github.com/windmill-labs/windmill/commit/fb12b23e0169ba2cdcf454a27dcf814a2caf26b3))
+
+## [1.752.0](https://github.com/windmill-labs/windmill/compare/v1.751.0...v1.752.0) (2026-07-07)
+
+
+### Features
+
+* add fork_parent_workspace claim to OIDC tokens for fork workspaces ([#9987](https://github.com/windmill-labs/windmill/issues/9987)) ([7efeae2](https://github.com/windmill-labs/windmill/commit/7efeae26d821b10667b6e3edd220468f6ae48936))
+* add SQL migrations for data tables ([#9693](https://github.com/windmill-labs/windmill/issues/9693)) ([e47aeda](https://github.com/windmill-labs/windmill/commit/e47aedac0a4af40dd697d5fc4d54dd3c8efe9ab8))
+* **cli:** clarify fork-branch workspace auto-targeting in output ([#9988](https://github.com/windmill-labs/windmill/issues/9988)) ([88c2d0e](https://github.com/windmill-labs/windmill/commit/88c2d0e8e32c218787c01daed80ef41efc39dd11))
+* open runs/schedules pages from AI chat in session preview tabs ([#9976](https://github.com/windmill-labs/windmill/issues/9976)) ([4bb82ad](https://github.com/windmill-labs/windmill/commit/4bb82ad6cdb62eae7b69b1054714333e54558632))
+* **raw-apps:** runtime-error overlay + AI import-React instruction ([#9966](https://github.com/windmill-labs/windmill/issues/9966)) ([8df613b](https://github.com/windmill-labs/windmill/commit/8df613b4d2f88765f49cc988a894ca323c4ec4f7))
+* **sessions:** v2 unified sidebar with family/fork scoping and preview router ([#9816](https://github.com/windmill-labs/windmill/issues/9816)) ([9503190](https://github.com/windmill-labs/windmill/commit/95031903ebe223dc03b49a6bcd3e4ee67cefc4bb))
+* smooth bursty AI chat streaming with a typewriter reveal ([#9991](https://github.com/windmill-labs/windmill/issues/9991)) ([a6276b5](https://github.com/windmill-labs/windmill/commit/a6276b590082d06480434a8ea002c335ea1cfb59))
+* update base image to debian 13 (trixie) ([#9973](https://github.com/windmill-labs/windmill/issues/9973)) ([c5c1ead](https://github.com/windmill-labs/windmill/commit/c5c1eadeb18e509a98d1e787206c0438417683fc))
+
+
+### Bug Fixes
+
+* **ai-agent:** align agent_actions_success with agent_actions for mcp and websearch ([#9983](https://github.com/windmill-labs/windmill/issues/9983)) ([87f8d46](https://github.com/windmill-labs/windmill/commit/87f8d46aafffd5e88a336192c51e0c95ff2e6f18))
+* **ai:** flow writer builds approval steps as scripts, not identity ([#9985](https://github.com/windmill-labs/windmill/issues/9985)) ([6b01caa](https://github.com/windmill-labs/windmill/commit/6b01caaf26a4f0a08f643db4e22a70e27d0dc554))
+* clear old path asset usage when renaming a script ([#9979](https://github.com/windmill-labs/windmill/issues/9979)) ([927b8d0](https://github.com/windmill-labs/windmill/commit/927b8d064f693384978184992b8f8a1cd708e711))
+* **cli:** auto-derive cascade triggers in --local pipeline graph ([#9978](https://github.com/windmill-labs/windmill/issues/9978)) ([edfe7b4](https://github.com/windmill-labs/windmill/commit/edfe7b415af6670c5855b7a0b52db4c1f7781964))
+* **pipelines:** live materialize/dataset editing — stale graph, phantom drafts, stale Save-all deploys ([#9990](https://github.com/windmill-labs/windmill/issues/9990)) ([f7efb64](https://github.com/windmill-labs/windmill/commit/f7efb646bf1f2e132d1e3ff031b142383ae01c5e))
+* **sessions:** auto-rename regression + preview-panel and fork nits ([#9993](https://github.com/windmill-labs/windmill/issues/9993)) ([804178f](https://github.com/windmill-labs/windmill/commit/804178f5e1c904c3f8e35e2b660f33c78964c6eb))
+* **sessions:** scope fork session Edits to session-edited items only ([#9989](https://github.com/windmill-labs/windmill/issues/9989)) ([7046dc6](https://github.com/windmill-labs/windmill/commit/7046dc6dfb474ef49313377855bb2bd60294e25a))
+
+## [1.751.0](https://github.com/windmill-labs/windmill/compare/v1.750.0...v1.751.0) (2026-07-06)
+
+
+### Features
+
+* add cosmetic dev/staging label for dev workspaces ([#9959](https://github.com/windmill-labs/windmill/issues/9959)) ([fd8e64d](https://github.com/windmill-labs/windmill/commit/fd8e64d11fea3ffdb7858100c67cb2e9ca841ed6))
+* **auth:** add runtime NO_AUTH mode for authentication bypass ([#9962](https://github.com/windmill-labs/windmill/issues/9962)) ([91e1b08](https://github.com/windmill-labs/windmill/commit/91e1b087a206efb7189824b4184e1f3f4cda7211))
+* **frontend:** custom skills — detail modal, batch manage, shared validation ([#9847](https://github.com/windmill-labs/windmill/issues/9847)) ([2e14302](https://github.com/windmill-labs/windmill/commit/2e14302e4abbad595584806bff12548d520fcb58))
+* **pipelines:** auto-derive cascade edges from ducklake/s3 reads (+ muted-read badge) ([#9963](https://github.com/windmill-labs/windmill/issues/9963)) ([3dcd394](https://github.com/windmill-labs/windmill/commit/3dcd3949a14199b106506994ea31ca3de7e636b3))
+
+
+### Bug Fixes
+
+* **ai:** centralize Anthropic Messages API routing across completion paths ([#9960](https://github.com/windmill-labs/windmill/issues/9960)) ([cc2f638](https://github.com/windmill-labs/windmill/commit/cc2f638de6cebeffb9fee1d4835a0cfd565af86c))
+* **assets:** responsive layout for small screens ([#9961](https://github.com/windmill-labs/windmill/issues/9961)) ([45946d1](https://github.com/windmill-labs/windmill/commit/45946d1185c0bd07948d4d8454880c2801571f9d))
+* **cli:** quote non-identifier property names in resource-type namespace ([#9964](https://github.com/windmill-labs/windmill/issues/9964)) ([dc6b997](https://github.com/windmill-labs/windmill/commit/dc6b99775b550e7433fee8a159c30eaf296500c5))
+* critical alerts modal mute toggles no longer close popover or fail to save ([#9969](https://github.com/windmill-labs/windmill/issues/9969)) ([6587019](https://github.com/windmill-labs/windmill/commit/6587019d263374ee5707d258f5d8eec7e73c690d))
+* **frontend:** theme-aware code block background in prose markdown ([#9968](https://github.com/windmill-labs/windmill/issues/9968)) ([9821596](https://github.com/windmill-labs/windmill/commit/9821596251cff698958ffbfbd11fffa6a7988c6c))
+
+## [1.750.0](https://github.com/windmill-labs/windmill/compare/v1.749.0...v1.750.0) (2026-07-06)
+
+
+### Features
+
+* chat-scoped session changes bar + unified diff drawer ([#9762](https://github.com/windmill-labs/windmill/issues/9762)) ([a6c0b37](https://github.com/windmill-labs/windmill/commit/a6c0b3756be78ca3fadc7bad6bae98c0887fd538))
+* **pipelines:** require data uploads before running a pipeline ([#9953](https://github.com/windmill-labs/windmill/issues/9953)) ([a1c5b7a](https://github.com/windmill-labs/windmill/commit/a1c5b7aa3ed2841f09f4f148ded5c5b5ef10fd3d))
+* **pipelines:** wm_partition macro for grain-agnostic partition filters ([#9950](https://github.com/windmill-labs/windmill/issues/9950)) ([43044c2](https://github.com/windmill-labs/windmill/commit/43044c2e28139b1dbde6844c781a821f8de68f58))
+
+
+### Bug Fixes
+
+* **ai:** test key routes Azure Foundry Claude models via Anthropic Messages API ([#9956](https://github.com/windmill-labs/windmill/issues/9956)) ([ea19cc9](https://github.com/windmill-labs/windmill/commit/ea19cc9dc459bd259e27f7fcc29601a010c5f8f0))
+* **cli:** HD-1 test_edges + HD-2 scd2 _current write in --local pipeline graph ([#9947](https://github.com/windmill-labs/windmill/issues/9947)) ([ad6f23d](https://github.com/windmill-labs/windmill/commit/ad6f23d6bfcf1056bcb6d8c6b552114e88177328))
+* **pipelines:** make node & pipeline-level run affordances always visible ([#9948](https://github.com/windmill-labs/windmill/issues/9948)) ([6eabb96](https://github.com/windmill-labs/windmill/commit/6eabb96ae78fb966f9916f907bb693d569b04c0b))
+* read chat drafts via own-draft route so drawer-kind drafts deploy ([#9913](https://github.com/windmill-labs/windmill/issues/9913)) ([056ebdb](https://github.com/windmill-labs/windmill/commit/056ebdb03543a93094c80ca354c117236cd8d6c8))
+* resolve extensionless bun relative imports on windows loader ([#9949](https://github.com/windmill-labs/windmill/issues/9949)) ([bf96621](https://github.com/windmill-labs/windmill/commit/bf9662172ad7e0ff53d39adc338fd7886672c8f9))
+
+## [1.749.0](https://github.com/windmill-labs/windmill/compare/v1.748.0...v1.749.0) (2026-07-05)
+
+
+### Features
+
+* **pipelines:** mid-DAG selective execution (dbt `model+`) for pipeline runs ([#9945](https://github.com/windmill-labs/windmill/issues/9945)) ([2d3a773](https://github.com/windmill-labs/windmill/commit/2d3a77344104a587548f23f1b614ceffd52a5778))
+* **pipelines:** partition run-arg picker + first-run setup signpost ([#9943](https://github.com/windmill-labs/windmill/issues/9943)) ([475b072](https://github.com/windmill-labs/windmill/commit/475b072987b33d50111f5251a5f69f4245f930ae))
+* **pipelines:** self-teaching custom data_test errors + scaffold ([#9937](https://github.com/windmill-labs/windmill/issues/9937)) ([0ad174f](https://github.com/windmill-labs/windmill/commit/0ad174fa490e17eeb26280b5bdfd62956dfed9ff))
+
+
+### Bug Fixes
+
+* **cli:** macro-library parity in --local pipeline graph + read-only run --dry-run ([#9942](https://github.com/windmill-labs/windmill/issues/9942)) ([e3f4303](https://github.com/windmill-labs/windmill/commit/e3f43033cafcdb5df253aeb55ce93e599b2584d2))
+* **datatable:** self-teaching error for unresolved datatable:// references ([#9941](https://github.com/windmill-labs/windmill/issues/9941)) ([55451db](https://github.com/windmill-labs/windmill/commit/55451db009e3060c21948ece2c97e102a3c9b171))
+* **object-storage:** remove 20-file bucket-browser listing cap in CE ([#9935](https://github.com/windmill-labs/windmill/issues/9935)) ([22452ce](https://github.com/windmill-labs/windmill/commit/22452ce54034a9bea8f7d48946818fd148b938c0))
+* **pipelines:** dedup guard for keyed merge + deploy-time SCD2 validation ([#9936](https://github.com/windmill-labs/windmill/issues/9936)) ([52ce805](https://github.com/windmill-labs/windmill/commit/52ce805f619747af4f998cde7819a164c754205a))
+* **pipelines:** link SCD2 &lt;dim&gt;_current view to its producer across all graph surfaces ([#9933](https://github.com/windmill-labs/windmill/issues/9933)) ([574d3ac](https://github.com/windmill-labs/windmill/commit/574d3ac9ff5015b5d3f53040c9d4dfbfd161a076))
+* **pipelines:** order data_test relationships refs before the tested script in a cascade ([#9934](https://github.com/windmill-labs/windmill/issues/9934)) ([46be39d](https://github.com/windmill-labs/windmill/commit/46be39dfb7fbfb2b70e61819d6065b45810c41c9))
+* **pipelines:** pipeline-level run control, tables label, data-test rollback + fork badges ([#9944](https://github.com/windmill-labs/windmill/issues/9944)) ([6ae8dd3](https://github.com/windmill-labs/windmill/commit/6ae8dd37b1de930ab17344cebf7c28385c6cfdba))
+* rebuild windows bun loader main.ts filter from forward-slash cdir ([#9946](https://github.com/windmill-labs/windmill/issues/9946)) ([a582e04](https://github.com/windmill-labs/windmill/commit/a582e04bf40cf685f88bceaf88e3d24bde3d420a))
+
+## [1.748.0](https://github.com/windmill-labs/windmill/compare/v1.747.0...v1.748.0) (2026-07-05)
+
+
+### Features
+
+* **ai-agent:** support reasoning effort in AI agent workflow steps ([#9886](https://github.com/windmill-labs/windmill/issues/9886)) ([a368d49](https://github.com/windmill-labs/windmill/commit/a368d49bd8786a2dca6771f2051f1d44d1b2363d))
+* **ducklake:** scheduled lake maintenance (expiry, compaction, orphan cleanup) ([#9916](https://github.com/windmill-labs/windmill/issues/9916)) ([3352150](https://github.com/windmill-labs/windmill/commit/33521505dbc34f22b575d21fda1cc76d698a8840))
+* **pipelines:** asset freshness — fresh/stale badge (CE) + watchdog (EE) ([#9909](https://github.com/windmill-labs/windmill/issues/9909)) ([5d7fb6d](https://github.com/windmill-labs/windmill/commit/5d7fb6deca3e02e89d77e5d3856483beb8b8bfeb))
+* **pipelines:** capture violating-row samples for data tests ([#9919](https://github.com/windmill-labs/windmill/issues/9919)) ([d4b4374](https://github.com/windmill-labs/windmill/commit/d4b4374de8f8a7875b050c16d1236fcd0355812b))
+* **pipelines:** fork data environments for ducklake materialization (dev data) ([#9915](https://github.com/windmill-labs/windmill/issues/9915)) ([39eb9de](https://github.com/windmill-labs/windmill/commit/39eb9de1bce400109c130a081807e40e995ae068))
+* **pipelines:** on_schema_change write guardrails + data_test deploy validation ([#9930](https://github.com/windmill-labs/windmill/issues/9930)) ([377c02e](https://github.com/windmill-labs/windmill/commit/377c02ec47389e64b7ef5cbbae0de05df648266d))
+* **pipelines:** record upstream snapshot ids on cascade-dispatched jobs ([#9910](https://github.com/windmill-labs/windmill/issues/9910)) ([af36498](https://github.com/windmill-labs/windmill/commit/af36498432e643108308e1c03b5d986d0f0f8888))
+* **pipelines:** schema contracts — save-time consumer checks vs captured schemas ([#9917](https://github.com/windmill-labs/windmill/issues/9917)) ([42e11c6](https://github.com/windmill-labs/windmill/commit/42e11c6570b62ffaa86598438fa8ddf462c4035f))
+* **pipeline:** write-audit-publish for materialization data tests ([#9911](https://github.com/windmill-labs/windmill/issues/9911)) ([dce247c](https://github.com/windmill-labs/windmill/commit/dce247c6d2678a2c95bd728027e17ae3965638e2))
+* **sdk:** enforce s3:// URIs for string S3 params + ingestion (EL) docs ([#9912](https://github.com/windmill-labs/windmill/issues/9912)) ([5ad2de9](https://github.com/windmill-labs/windmill/commit/5ad2de91a26b312bf27124ceca16ef331621bde8))
+
+
+### Bug Fixes
+
+* **cli:** pipeline + workspace UX batch (init/bind stub, run errors, macro libs, lock-job report, upgrade errors) ([#9929](https://github.com/windmill-labs/windmill/issues/9929)) ([28a6b08](https://github.com/windmill-labs/windmill/commit/28a6b086c842105298f236baa0a61868f71a5eb1))
+* **cli:** publish all windmill-parser-wasm-* deps so local pipeline graph keeps write edges ([#9926](https://github.com/windmill-labs/windmill/issues/9926)) ([744a759](https://github.com/windmill-labs/windmill/commit/744a7597edaf3ca9a7fd2b21a34fb33457913a64))
+* **pipelines:** activity-axis label clarity + select failed node on cascade failure ([#9931](https://github.com/windmill-labs/windmill/issues/9931)) ([5769b60](https://github.com/windmill-labs/windmill/commit/5769b6036cf14b0cb424c5b3d9d878c600a5652e))
+
+## [1.747.0](https://github.com/windmill-labs/windmill/compare/v1.746.0...v1.747.0) (2026-07-03)
+
+
+### Features
+
+* **frontend:** add federatedTokenFile field to instance object storage Azure config ([#9904](https://github.com/windmill-labs/windmill/issues/9904)) ([ae85d27](https://github.com/windmill-labs/windmill/commit/ae85d274371a24c5badb6081f00deeb409123252))
+
+
+### Bug Fixes
+
+* **ai:** route Azure Foundry Claude models via Anthropic Messages API ([#9908](https://github.com/windmill-labs/windmill/issues/9908)) ([d600c7e](https://github.com/windmill-labs/windmill/commit/d600c7ecfe305533798e82e8d05e5f2f297f9b54))
+* **forks:** clone only the current raw-app bundle, via server-side copy ([#9899](https://github.com/windmill-labs/windmill/issues/9899)) ([5c521d8](https://github.com/windmill-labs/windmill/commit/5c521d808a2b5d6d6bb7cf3da17fb2addc53fdf4))
+* **kafka:** set https.ca.location=probe for OAUTHBEARER OIDC token endpoint ([#9897](https://github.com/windmill-labs/windmill/issues/9897)) ([1b6065f](https://github.com/windmill-labs/windmill/commit/1b6065fa9201fd548c4b2ef199f1009200645929))
+* prevent truncated tool call args from bricking AI chat sessions ([#9902](https://github.com/windmill-labs/windmill/issues/9902)) ([4ba17d0](https://github.com/windmill-labs/windmill/commit/4ba17d0f9cd70489f89c84f982a0c8f0062fed1a))
+* strip NUL characters from app values at save time ([#9903](https://github.com/windmill-labs/windmill/issues/9903)) ([3ec1f16](https://github.com/windmill-labs/windmill/commit/3ec1f164be9c8c6c40e003188ce593a963c65a43))
+
+## [1.746.0](https://github.com/windmill-labs/windmill/compare/v1.745.0...v1.746.0) (2026-07-02)
+
+
+### Features
+
+* **ai:** add Azure AI Foundry as a native AI provider ([#9879](https://github.com/windmill-labs/windmill/issues/9879)) ([d9b080f](https://github.com/windmill-labs/windmill/commit/d9b080f57fa0be144cefa773d39742c45b40f043))
+* **frontend:** group compare & deploy items by folder ([#9880](https://github.com/windmill-labs/windmill/issues/9880)) ([7b04820](https://github.com/windmill-labs/windmill/commit/7b04820f8ef8c7f02f79dd4239a877f667d23e6a))
+* **frontend:** pipelines index page and sql editor hint ([#9881](https://github.com/windmill-labs/windmill/issues/9881)) ([20351a6](https://github.com/windmill-labs/windmill/commit/20351a6b4c262184c5f815eeb5de007ab1eaf4a0))
+* **pipeline:** backfill a range of partitions from the asset drawer ([#9885](https://github.com/windmill-labs/windmill/issues/9885)) ([53bbb92](https://github.com/windmill-labs/windmill/commit/53bbb92953178eb6d0017818ef870f3cb2399dfd))
+* **pipelines:** workspace duckdb macro libraries (// macros / // use) ([#9890](https://github.com/windmill-labs/windmill/issues/9890)) ([84141ad](https://github.com/windmill-labs/windmill/commit/84141add1ddf35c2573e3c213366ce7c5f1f2258))
+* **s3:** replace CE 50MB upload cap with 10GiB workspace storage quota ([#9874](https://github.com/windmill-labs/windmill/issues/9874)) ([af01e90](https://github.com/windmill-labs/windmill/commit/af01e90b5c65d1b1cfacf4433f8cff7effe73768))
+* support workspace forks on cloud using parent workspace limits ([#9864](https://github.com/windmill-labs/windmill/issues/9864)) ([7c7d747](https://github.com/windmill-labs/windmill/commit/7c7d7474cc86a4052272032f281cc4d7a85db37b))
+
+
+### Bug Fixes
+
+* **duckdb:** auto-declare partition arg for `// partitioned` scripts ([#9878](https://github.com/windmill-labs/windmill/issues/9878)) ([b883adb](https://github.com/windmill-labs/windmill/commit/b883adbc0011073da592dc5b39e1b79db492c83c))
+* **frontend:** home New submenus fall back below, hugging the right edge ([#9894](https://github.com/windmill-labs/windmill/issues/9894)) ([186ac49](https://github.com/windmill-labs/windmill/commit/186ac4933b79aed57fce23ebcf3b525fcfd1c474))
+* **frontend:** show inline workspace name editor on general settings (Fixes GIT-911) ([#9892](https://github.com/windmill-labs/windmill/issues/9892)) ([a49c087](https://github.com/windmill-labs/windmill/commit/a49c0871d7ab2aaf78a7713b8a786ead937434da))
+* **frontend:** stack cron field and cron builder button on narrow screens ([#9871](https://github.com/windmill-labs/windmill/issues/9871)) ([7989795](https://github.com/windmill-labs/windmill/commit/79897950e7646b00d92a28a009174d91c705b251))
+* invalidate bun bundle cache on transitive relative-import changes ([#9891](https://github.com/windmill-labs/windmill/issues/9891)) ([d15033c](https://github.com/windmill-labs/windmill/commit/d15033cde6a474b548ebbaf18ff02223fc21f701))
+* make SMTP username and password optional in frontend validation ([#9895](https://github.com/windmill-labs/windmill/issues/9895)) ([37bb574](https://github.com/windmill-labs/windmill/commit/37bb57474e8336823bb31527f2a708ef41cd39c4))
+* **parsers:** infer py s3 assets from S3Object constructor and dict forms ([#9877](https://github.com/windmill-labs/windmill/issues/9877)) ([659642e](https://github.com/windmill-labs/windmill/commit/659642e4889361f86e8addb038cda62fc3471006))
+* pipeline dogfooding fixes — SCD2 data-test scope, --partition, s3object upload binding ([#9875](https://github.com/windmill-labs/windmill/issues/9875)) ([d65f58c](https://github.com/windmill-labs/windmill/commit/d65f58c388d88fff71cda22dfa21aecdae70c450))
+* polish pipeline graph view (layout, viewport, minimap, lineage, timestamps) ([#9883](https://github.com/windmill-labs/windmill/issues/9883)) ([b92a86b](https://github.com/windmill-labs/windmill/commit/b92a86b8b3a60b877540c3a7f0ffefe36ccbb053))
+* stale AI chat context picker after workspace item changes ([#9893](https://github.com/windmill-labs/windmill/issues/9893)) ([5af91a6](https://github.com/windmill-labs/windmill/commit/5af91a677cad88faccba702e3556fc4fb7b6e640))
+* **triggers:** retry transient websocket connect failures before disabling ([#9887](https://github.com/windmill-labs/windmill/issues/9887)) ([7894507](https://github.com/windmill-labs/windmill/commit/789450731b0a3c8dffa336f7bfc3f3de528c09fb))
+
+## [1.745.0](https://github.com/windmill-labs/windmill/compare/v1.744.0...v1.745.0) (2026-07-01)
+
+
+### Features
+
+* **forks:** partial-visibility deploy + surface hidden items ([#9868](https://github.com/windmill-labs/windmill/issues/9868)) ([20cd1a0](https://github.com/windmill-labs/windmill/commit/20cd1a02d582c0715bedacce52cc5c1e1e8d70ca))
+* **frontend:** add zoom and download to Mermaid graphs ([#9859](https://github.com/windmill-labs/windmill/issues/9859)) ([289017b](https://github.com/windmill-labs/windmill/commit/289017bcb28c049c8258b2ffd7da0ec3e6ef120b))
+* use derived username instead of email for non-member superadmins ([#9857](https://github.com/windmill-labs/windmill/issues/9857)) ([76a9523](https://github.com/windmill-labs/windmill/commit/76a95230095ca3f43c9dc9eecde0e9de6520242f))
+
+
+### Bug Fixes
+
+* **cli:** correct misleading delete-fork command description ([#9870](https://github.com/windmill-labs/windmill/issues/9870)) ([a73b14d](https://github.com/windmill-labs/windmill/commit/a73b14d902d759226d0af2f2faf9bdd6588e358c))
+* **folders:** allow dots and at-signs in folder owner validation ([#9856](https://github.com/windmill-labs/windmill/issues/9856)) ([383c705](https://github.com/windmill-labs/windmill/commit/383c70523bf81c5c07784a4379ef6b1c93ff86e5))
+* **forks:** require admin of both sides for the compare visibility guard ([#9869](https://github.com/windmill-labs/windmill/issues/9869)) ([7363d2c](https://github.com/windmill-labs/windmill/commit/7363d2c217cb04391f03b2f9958be70a9d0b5325))
+* **forks:** reset diff tally on trigger delete + guard compare visibility for admins ([#9866](https://github.com/windmill-labs/windmill/issues/9866)) ([6a6f129](https://github.com/windmill-labs/windmill/commit/6a6f12960e29c314d11ad541519c71412f42567b))
+* **jobs:** give flow dynselect a path and its worker tag, like scripts ([#9867](https://github.com/windmill-labs/windmill/issues/9867)) ([1a9debb](https://github.com/windmill-labs/windmill/commit/1a9debb689f756f38db085d1360c8fe7691ada48))
+* **offboarding:** make global reassignment per-workspace and optional ([#9863](https://github.com/windmill-labs/windmill/issues/9863)) ([3586164](https://github.com/windmill-labs/windmill/commit/35861641f807a02b5c205608fb592e20ee7cad7f))
+
+## [1.744.0](https://github.com/windmill-labs/windmill/compare/v1.743.0...v1.744.0) (2026-07-01)
+
+
+### Features
+
+* add copy-to-clipboard button to rendered Mermaid diagrams in AI chat ([#9838](https://github.com/windmill-labs/windmill/issues/9838)) ([a27e814](https://github.com/windmill-labs/windmill/commit/a27e814a03c615259381eaf684aa90d56569b0af))
+* add dev workspaces paired with a lockable prod workspace ([#9793](https://github.com/windmill-labs/windmill/issues/9793)) ([b4b0c6a](https://github.com/windmill-labs/windmill/commit/b4b0c6a93e52152251fadefe319773faf42549b2))
+* **ansible:** support repo-provided ansible.cfg in delegate_to_git_repo ([#9851](https://github.com/windmill-labs/windmill/issues/9851)) ([68bf0da](https://github.com/windmill-labs/windmill/commit/68bf0daf5815307cda6ce23214dd5159b6aa33b4))
+* **licensing:** enforce offline license seat cap ([#9845](https://github.com/windmill-labs/windmill/issues/9845)) ([83f3d7f](https://github.com/windmill-labs/windmill/commit/83f3d7f910b331c09f60cc9ff556728afa3dec07))
+* **object-store:** make GCS service account key optional for Workload Identity ([#9842](https://github.com/windmill-labs/windmill/issues/9842)) ([83ed011](https://github.com/windmill-labs/windmill/commit/83ed011e264f20ffa66a7bf933f2fe3615cf6b67))
+* **pipeline:** local development for data pipelines (CLI --local + pipeline dev preview) ([#9840](https://github.com/windmill-labs/windmill/issues/9840)) ([74f579e](https://github.com/windmill-labs/windmill/commit/74f579e6d9ef08e74460f904a4c22ed9d6a3b5b0))
+* **pipelines:** add managed SCD2 history materialize strategy ([#9850](https://github.com/windmill-labs/windmill/issues/9850)) ([5a66127](https://github.com/windmill-labs/windmill/commit/5a661279a3690e2393b9b16996f5d1a5a509259c))
+
+
+### Bug Fixes
+
+* **ai-chat:** replay anthropic turns verbatim to keep thinking valid ([#9843](https://github.com/windmill-labs/windmill/issues/9843)) ([a37a144](https://github.com/windmill-labs/windmill/commit/a37a144e81cf6b3de935688a617e9d0e1756004a))
+* grant dispatch_event table to windmill roles ([#9852](https://github.com/windmill-labs/windmill/issues/9852)) ([f05b50d](https://github.com/windmill-labs/windmill/commit/f05b50d29ac2fdbb808a97057fb92c8e425b4a2f))
+* grant workspace_diff, materialized_partition, debounce_stale_data to windmill roles ([#9853](https://github.com/windmill-labs/windmill/issues/9853)) ([293647d](https://github.com/windmill-labs/windmill/commit/293647de4c13cb8468cbd81ff1924cba90e164b4))
+* honor verify-ca/verify-full sslmode for postgres connections ([#9835](https://github.com/windmill-labs/windmill/issues/9835)) ([bf6be96](https://github.com/windmill-labs/windmill/commit/bf6be967fa8c74e1299cf63f813c1cfa34b97f3e))
+* **mcp:** stop double-escaping string query params in build_query_string ([#9855](https://github.com/windmill-labs/windmill/issues/9855)) ([1c46f89](https://github.com/windmill-labs/windmill/commit/1c46f899ca03edf62053f4f14d65b4eabff4255d))
+* **s3_proxy:** preserve URL-encoding on Hive-partition proxy writes ([#9848](https://github.com/windmill-labs/windmill/issues/9848)) ([6b79bdd](https://github.com/windmill-labs/windmill/commit/6b79bddd42fe55f891c17cb71a7e36ee31337bac))
+* validate workspace name length (max 50 chars) on create and fork ([#9854](https://github.com/windmill-labs/windmill/issues/9854)) ([b52972d](https://github.com/windmill-labs/windmill/commit/b52972d0de89004e98d18241d238ca028e4eecba))
+
+## [1.743.0](https://github.com/windmill-labs/windmill/compare/v1.742.0...v1.743.0) (2026-06-29)
+
+
+### Features
+
+* **home:** redesign create-new popover and home header ([#9827](https://github.com/windmill-labs/windmill/issues/9827)) ([2493eaf](https://github.com/windmill-labs/windmill/commit/2493eaf031f30072637a297398674e761f039005))
+* **pipeline:** AI-chat data-pipeline editor (route + in-session) + home surfacing ([#9805](https://github.com/windmill-labs/windmill/issues/9805)) ([c910278](https://github.com/windmill-labs/windmill/commit/c91027824be1f1f49cdd14148baf6aad092a1dd0))
+
+
+### Bug Fixes
+
+* **gcp:** require token verification for authenticated push delivery ([#9834](https://github.com/windmill-labs/windmill/issues/9834)) ([9b65161](https://github.com/windmill-labs/windmill/commit/9b65161c643bf3f120d2ebd82f786c17233a971b))
+
+## [1.742.0](https://github.com/windmill-labs/windmill/compare/v1.741.0...v1.742.0) (2026-06-28)
+
+
+### Features
+
+* **apps:** add labels input to app editor deploy drawer ([#9828](https://github.com/windmill-labs/windmill/issues/9828)) ([da45e69](https://github.com/windmill-labs/windmill/commit/da45e699c8aefeede172c90769ef4f4b182fec0c))
+* column-level lineage for DuckLake pipelines (SQL-AST inferred + traceable) ([#9814](https://github.com/windmill-labs/windmill/issues/9814)) ([003a262](https://github.com/windmill-labs/windmill/commit/003a262a4e9d6c2a63ada01aa8429aea1fbb6031))
+
+
+### Bug Fixes
+
+* **audit:** don't read pg_authid from an elevated context in S3 export migration ([#9832](https://github.com/windmill-labs/windmill/issues/9832)) ([75ba81b](https://github.com/windmill-labs/windmill/commit/75ba81b2d27fb0722095780312064cb93d20287e))
+* close unauthenticated DAP debugger program-mode launch bypass ([#9829](https://github.com/windmill-labs/windmill/issues/9829)) ([c0768de](https://github.com/windmill-labs/windmill/commit/c0768de0acdf63eaba5fb97d04bfc64f2f03b93d))
+* redeploy older app version from deployment history ([#9826](https://github.com/windmill-labs/windmill/issues/9826)) ([c479afa](https://github.com/windmill-labs/windmill/commit/c479afab8ebceccbee050e923dc5c27a6712ea62))
+
+## [1.741.0](https://github.com/windmill-labs/windmill/compare/v1.740.0...v1.741.0) (2026-06-26)
+
+
+### Features
+
+* **ai-chat:** add create_folder tool to global chat ([#9819](https://github.com/windmill-labs/windmill/issues/9819)) ([44c25de](https://github.com/windmill-labs/windmill/commit/44c25de418612ab98341adb15d5671222b54367e))
+* **ai-chat:** hint /compact in context usage tooltip ([#9777](https://github.com/windmill-labs/windmill/issues/9777)) ([aadfb62](https://github.com/windmill-labs/windmill/commit/aadfb620c0b7dcd7e94367b761875e14ef9abe69))
+* **ai-chat:** let global chat edit the user's personal instructions ([#9771](https://github.com/windmill-labs/windmill/issues/9771)) ([3be2752](https://github.com/windmill-labs/windmill/commit/3be27521b05de33e48582e80c6651071f889f048))
+* **ai-chat:** surface raw apps in the @-mention context picker ([#9800](https://github.com/windmill-labs/windmill/issues/9800)) ([1602244](https://github.com/windmill-labs/windmill/commit/16022447c7b445be753b9545b10b4c67da0893d5))
+* capture managed-materialize output schema as asset metadata ([#2](https://github.com/windmill-labs/windmill/issues/2)a) ([#9812](https://github.com/windmill-labs/windmill/issues/9812)) ([ade74b2](https://github.com/windmill-labs/windmill/commit/ade74b297f6a03441e700a20ffc2d7291c8a85fd))
+* **sdk:** allow overriding worker tag when running jobs (WIN-2105) ([#9807](https://github.com/windmill-labs/windmill/issues/9807)) ([52fc7bf](https://github.com/windmill-labs/windmill/commit/52fc7bf94cf3f87f68d9dba9884944d87e7d5d57))
+
+
+### Bug Fixes
+
+* apply step timeout to 'Test this step' preview ([#9810](https://github.com/windmill-labs/windmill/issues/9810)) ([d04062b](https://github.com/windmill-labs/windmill/commit/d04062bff58c9c4c79ce542a4321e71bcbcf0e98))
+* **flows:** reject corrupt step paths at deploy + atomic cache writes ([#9751](https://github.com/windmill-labs/windmill/issues/9751)) ([#9813](https://github.com/windmill-labs/windmill/issues/9813)) ([3cda447](https://github.com/windmill-labs/windmill/commit/3cda44762148bcd2ee5c0ea821db884950376ead))
+* **frontend:** clarify instance data table unavailable on cloud ([#9806](https://github.com/windmill-labs/windmill/issues/9806)) ([c3e8c78](https://github.com/windmill-labs/windmill/commit/c3e8c789ac05c9c28991d9ab6f2358f61fa87971))
+* hide GCS service account key behind a reveal in object storage settings ([#9815](https://github.com/windmill-labs/windmill/issues/9815)) ([0ec5061](https://github.com/windmill-labs/windmill/commit/0ec5061270749ed078e01f5a4bc7397a1755ca32))
+* ping job during volume setup to prevent false zombie restarts ([#9803](https://github.com/windmill-labs/windmill/issues/9803)) ([43bb676](https://github.com/windmill-labs/windmill/commit/43bb676dc5652cb06fe1414b8d3aacf295bae36b))
+* skipped suspend step no longer parks the flow forever ([#9821](https://github.com/windmill-labs/windmill/issues/9821)) ([40110bc](https://github.com/windmill-labs/windmill/commit/40110bc7158bc42c3d84bd4637a12b82fcd72a9a))
+
+
+### Performance Improvements
+
+* **audit:** re-anchor S3 audit export on enable + opt-in backfill ([#9818](https://github.com/windmill-labs/windmill/issues/9818)) ([577ceee](https://github.com/windmill-labs/windmill/commit/577ceeee8679f054c6898d1a7889df30ab830f8f))
+
+## [1.740.0](https://github.com/windmill-labs/windmill/compare/v1.739.0...v1.740.0) (2026-06-25)
+
+
+### Features
+
+* **api:** add structured endpoint for flow logs ([#9797](https://github.com/windmill-labs/windmill/issues/9797)) ([ba768fe](https://github.com/windmill-labs/windmill/commit/ba768fee888682cb50142d6e76c0422c40307f46))
+* bounded-cascade selective execution for pipelines (UI + CLI) ([#9695](https://github.com/windmill-labs/windmill/issues/9695)) ([248540a](https://github.com/windmill-labs/windmill/commit/248540ac4d6e4ee9ee7c3e6f2cc822c63cc6426e))
+* data tests for ducklake pipeline materialization ([#9708](https://github.com/windmill-labs/windmill/issues/9708)) ([f6998ec](https://github.com/windmill-labs/windmill/commit/f6998ec54cba2507703790bf33427e7567d42c4b))
+* detect and guard against deploying stale drafts ([#9768](https://github.com/windmill-labs/windmill/issues/9768)) ([d865518](https://github.com/windmill-labs/windmill/commit/d8655189347f58df9d17e83dc55798baf7964279))
+* ducklake time-travel UX (snapshot history + AT VERSION reads) ([#9709](https://github.com/windmill-labs/windmill/issues/9709)) ([d131d75](https://github.com/windmill-labs/windmill/commit/d131d754e1fc9674abf5de383d2bc93596df9bd1))
+* self-host docs search for chat, mcp, cli; drop inkeep ([#9772](https://github.com/windmill-labs/windmill/issues/9772)) ([9d61e4e](https://github.com/windmill-labs/windmill/commit/9d61e4e59e4101de84217f7c7846f1aa94e84d89))
+
+
+### Bug Fixes
+
+* allow hyphens in postgresql database name validation ([#9782](https://github.com/windmill-labs/windmill/issues/9782)) ([170cd79](https://github.com/windmill-labs/windmill/commit/170cd79aaf92152fc3c0f675f155853c7f0e5b25))
+* **debounce:** never supersede a running debounce survivor ([#9780](https://github.com/windmill-labs/windmill/issues/9780)) ([5549bdc](https://github.com/windmill-labs/windmill/commit/5549bdc67a5559a764616c44b1018543bc0568fe))
+* decrypt secret variables via external backend in common resolvers ([#9784](https://github.com/windmill-labs/windmill/issues/9784)) ([cd42c6c](https://github.com/windmill-labs/windmill/commit/cd42c6ca18261328055554788932c3fe876a4a5b))
+* enforce containment of python module dir for preview jobs ([#9704](https://github.com/windmill-labs/windmill/issues/9704)) ([88fca6a](https://github.com/windmill-labs/windmill/commit/88fca6a8c130b9e3b0f0cd410e422d4e074fc11f))
+* **frontend:** apply script editor timeout to preview/Test runs ([#9794](https://github.com/windmill-labs/windmill/issues/9794)) ([6664ce6](https://github.com/windmill-labs/windmill/commit/6664ce6dc0c5fbc283303148de06d6bb85e4acf7))
+* **frontend:** nested-loop "Test this step" resolves iter to innermost loop ([#9778](https://github.com/windmill-labs/windmill/issues/9778)) ([74ebfc6](https://github.com/windmill-labs/windmill/commit/74ebfc67f069047875db738926865bd4bd6fe9e9))
+* opt out of Deno minimum-dependency-age for private npm registries ([#9802](https://github.com/windmill-labs/windmill/issues/9802)) ([b28f974](https://github.com/windmill-labs/windmill/commit/b28f974e5069f635419d9ea56fad6a0e417894e8))
+* pass SSL cert env vars to `uv python install` ([#9790](https://github.com/windmill-labs/windmill/issues/9790)) ([962758c](https://github.com/windmill-labs/windmill/commit/962758c02de5f6d962c681fe9c39769b99429e8d))
+* **python:** re-verify wheel RECORD on local cache reuse (once per worker) ([#9775](https://github.com/windmill-labs/windmill/issues/9775)) ([6c71c33](https://github.com/windmill-labs/windmill/commit/6c71c33470e3ea547f3b994db829eb4d04882443))
+* **python:** serialize concurrent installs into shared wheel cache dir ([#9787](https://github.com/windmill-labs/windmill/issues/9787)) ([11d83ab](https://github.com/windmill-labs/windmill/commit/11d83ab1ec559be5d3263010228e6db65358e04b))
+* re-pin stale-draft fork base when restoring an app deployment ([#9792](https://github.com/windmill-labs/windmill/issues/9792)) ([b9711e5](https://github.com/windmill-labs/windmill/commit/b9711e5ace8585315a1c2b85bb25ac8dd7832d6f))
+* restore libargon2-1 for PHP runtime in server image ([#9795](https://github.com/windmill-labs/windmill/issues/9795)) ([e9cb806](https://github.com/windmill-labs/windmill/commit/e9cb80639b2dec63ede69fc3a4e3720bb1a3c319))
+* use transaction for parallel_monitor_lock DELETE in last-iteration path ([#9789](https://github.com/windmill-labs/windmill/issues/9789)) ([754cae9](https://github.com/windmill-labs/windmill/commit/754cae956ac8d431ddeb055453835e249cdd07b7))
+
+
+### Performance Improvements
+
+* drop v2_job side-table ON DELETE CASCADE FKs to speed retention deletes ([#9786](https://github.com/windmill-labs/windmill/issues/9786)) ([aa098c7](https://github.com/windmill-labs/windmill/commit/aa098c70c0271b2b1917749d1f607c0559cf04de))
+* eliminate dual-connection DB pool contention across worker, queue, and api ([#9798](https://github.com/windmill-labs/windmill/issues/9798)) ([0dbd9c1](https://github.com/windmill-labs/windmill/commit/0dbd9c1231b00d4693af68835fe1d9e7c8869b43))
+
+## [1.739.0](https://github.com/windmill-labs/windmill/compare/v1.738.0...v1.739.0) (2026-06-24)
+
+
+### Features
+
+* add /compact session chat command ([#9764](https://github.com/windmill-labs/windmill/issues/9764)) ([83cc553](https://github.com/windmill-labs/windmill/commit/83cc5533ee92e59356a117eefeaf42dad23287f6))
+* add session chat slash commands ([#9748](https://github.com/windmill-labs/windmill/issues/9748)) ([24b95e9](https://github.com/windmill-labs/windmill/commit/24b95e9fe12ba4abdfe1ff6e9f9fe42cb2ded011))
+* **ai-chat:** add /clear session command to start a fresh conversation ([#9769](https://github.com/windmill-labs/windmill/issues/9769)) ([3fafac2](https://github.com/windmill-labs/windmill/commit/3fafac275d2100a6f89924040650cf959945d209))
+* **ai-chat:** context usage gauge + unified model settings menu ([#9763](https://github.com/windmill-labs/windmill/issues/9763)) ([2e020b2](https://github.com/windmill-labs/windmill/commit/2e020b2ccc7a649a5923bff72a98f07d4fc85381))
+* **apps:** show raw-app fork diffs as per-file tree items ([#9491](https://github.com/windmill-labs/windmill/issues/9491)) ([e98df38](https://github.com/windmill-labs/windmill/commit/e98df38ac43823ee85209a4b09cd70690469302d))
+* **frontend:** add filter submenu to collapsed AI sessions popover ([#9757](https://github.com/windmill-labs/windmill/issues/9757)) ([3d48ba7](https://github.com/windmill-labs/windmill/commit/3d48ba7738c3d3356539b5fc44a871f6b7f9d548))
+* **frontend:** restore raw app 'open preview in separate window' ([#9765](https://github.com/windmill-labs/windmill/issues/9765)) ([a116715](https://github.com/windmill-labs/windmill/commit/a116715c418c39d48a91e6c0b4484a31537dff38))
+* **frontend:** show approval wait as a distinct segment in flow timeline ([#9756](https://github.com/windmill-labs/windmill/issues/9756)) ([2a70ccc](https://github.com/windmill-labs/windmill/commit/2a70ccc38675c7c2353807a4f85764a8a35224e2))
+* scope AI sessions per workspace root with lifecycle reconcile ([#9734](https://github.com/windmill-labs/windmill/issues/9734)) ([42c5e7a](https://github.com/windmill-labs/windmill/commit/42c5e7a3fc9b74256de8806ec0d7b62e8bbf029c))
+
+
+### Bug Fixes
+
+* **ai-chat:** strip unclosed &lt;summary&gt; tag leaking into compaction summary ([#9750](https://github.com/windmill-labs/windmill/issues/9750)) ([250a05f](https://github.com/windmill-labs/windmill/commit/250a05f544ae397bb91af5fc83bf408cfe1c554d))
+* **apps:** realign legacy raw-app drafts to raw_app draft kind ([#9761](https://github.com/windmill-labs/windmill/issues/9761)) ([288318a](https://github.com/windmill-labs/windmill/commit/288318ac269714fc03b15622dbb86b1c28268a36))
+* **backend:** resolve folder_labels search_path on non-public (PG_SCHEMA) schemas ([#9758](https://github.com/windmill-labs/windmill/issues/9758)) ([f582878](https://github.com/windmill-labs/windmill/commit/f5828780fd6a8be070b2933ebd41ee6dff98a9e1))
+* forbid superadmin job tokens from global user and token management ([#9715](https://github.com/windmill-labs/windmill/issues/9715)) ([043c2c0](https://github.com/windmill-labs/windmill/commit/043c2c05b7678c49faca0ccb28e5f6393567ba4d))
+* **frontend:** highlight the runtime-chosen branch in flow graph viewer ([#9755](https://github.com/windmill-labs/windmill/issues/9755)) ([de6192b](https://github.com/windmill-labs/windmill/commit/de6192bec1695883a07452f7db2fb51c94dbfd43))
+* **frontend:** keep #content portal target present on AI-session route ([#9754](https://github.com/windmill-labs/windmill/issues/9754)) ([5e09c50](https://github.com/windmill-labs/windmill/commit/5e09c501713ebbe05b28ce0084eca641f0dbe95c))
+* **frontend:** show AI skills settings only when global mode enabled ([#9747](https://github.com/windmill-labs/windmill/issues/9747)) ([c017f7f](https://github.com/windmill-labs/windmill/commit/c017f7f8919a51292ddf01574961d1774bc1ba23))
+* **frontend:** stop flow step id generation from being poisoned by non-canonical keys ([#9766](https://github.com/windmill-labs/windmill/issues/9766)) ([4dbf873](https://github.com/windmill-labs/windmill/commit/4dbf8737238ccc4dc2c67365e6d43f04f46c75b5))
+* persist on-behalf-of user across app deploy paths ([#9773](https://github.com/windmill-labs/windmill/issues/9773)) ([f99781c](https://github.com/windmill-labs/windmill/commit/f99781ca5f77248206c951935cc44acfa5f072eb))
+* reject symlink traversal in job-dir path validation ([#9713](https://github.com/windmill-labs/windmill/issues/9713)) ([b5bd824](https://github.com/windmill-labs/windmill/commit/b5bd8245d81b84fc14d3ea955bf1e66ac576bf37))
+
+
+### Performance Improvements
+
+* **audit:** adaptive timestamp floor for S3 audit-log export ([#9752](https://github.com/windmill-labs/windmill/issues/9752)) ([55bed4a](https://github.com/windmill-labs/windmill/commit/55bed4abcfce2a611b16054573980d2eb613ccb3))
+* **monitor:** vacuum job_perms/job_result_stream right after each orphan sweep ([#9753](https://github.com/windmill-labs/windmill/issues/9753)) ([8912e21](https://github.com/windmill-labs/windmill/commit/8912e21d1571e57b5cf21b7d4d9520e20a28e70d))
+
+## [1.738.0](https://github.com/windmill-labs/windmill/compare/v1.737.0...v1.738.0) (2026-06-23)
+
+
+### Features
+
+* add resource and infrastructure telemetry ([#9737](https://github.com/windmill-labs/windmill/issues/9737)) ([9793d01](https://github.com/windmill-labs/windmill/commit/9793d01575415963a89609a1baf2cd64f0d050cc))
+* render mermaid diagrams in chat code blocks ([#9738](https://github.com/windmill-labs/windmill/issues/9738)) ([cfb9f1d](https://github.com/windmill-labs/windmill/commit/cfb9f1dbc23110ecf8f91bb3c8c81fc6e35dc09b))
+
+
+### Bug Fixes
+
+* **ai-chat:** Fix incorrect editor edits from ai chat [#1](https://github.com/windmill-labs/windmill/issues/1) ([#9741](https://github.com/windmill-labs/windmill/issues/9741)) ([fc797a3](https://github.com/windmill-labs/windmill/commit/fc797a35fe7885630c81453df0fc94769e73873a))
+* allow object storage test for non-super-admins, harden on cloud ([#9739](https://github.com/windmill-labs/windmill/issues/9739)) ([24446e8](https://github.com/windmill-labs/windmill/commit/24446e80093ade349f7fbf65063d2d1cb5551c1e))
+* **frontend:** debounce external code→Monaco sync in Editor ([#9743](https://github.com/windmill-labs/windmill/issues/9743)) ([29c67ce](https://github.com/windmill-labs/windmill/commit/29c67ced97bf2919584986f9d9eceb4337c34ad9))
+* **frontend:** preserve editor content when closing instance settings drawer ([#9740](https://github.com/windmill-labs/windmill/issues/9740)) ([11d0e65](https://github.com/windmill-labs/windmill/commit/11d0e65f3af9a048bc1921bbdd3d676a07483a57))
+* pipeline annotation false-positives from body comments ([#9736](https://github.com/windmill-labs/windmill/issues/9736)) ([984ea72](https://github.com/windmill-labs/windmill/commit/984ea728d98649b66b1cae899bdab9af3176caa7))
+* preserve fork parent linkage on workspace id change ([#9716](https://github.com/windmill-labs/windmill/issues/9716)) ([cbf54d4](https://github.com/windmill-labs/windmill/commit/cbf54d4eb432638e27f67c4c8b879cbcc0291da3))
+* prevent variable push from corrupting is_secret variables ([#9705](https://github.com/windmill-labs/windmill/issues/9705)) ([ba4b368](https://github.com/windmill-labs/windmill/commit/ba4b368706e95e22f346a10e5fe145b0795ac3f6))
+
+
+### Performance Improvements
+
+* **monitor:** skip protected prefix in retention delete via cross-batch watermark (WIN-2088) ([#9744](https://github.com/windmill-labs/windmill/issues/9744)) ([e90b2be](https://github.com/windmill-labs/windmill/commit/e90b2be8fade1eb78cd685890291f5a4553a6a10))
+
 ## [1.737.0](https://github.com/windmill-labs/windmill/compare/v1.736.0...v1.737.0) (2026-06-23)


@@ -1,18 +1,27 @@
 {
 	layer4 {
 		:25 {
-			proxy {
-				to windmill_server:2525
+			route {
+				proxy {
+					upstream windmill_server:2525
+				}
 			}
 		}
 	}
 }

 {$BASE_URL} {
-        bind {$ADDRESS}
+        # Default to all interfaces (IPv4 + IPv6) when ADDRESS is unset. A bare
+        # `bind {$ADDRESS}` with an empty value makes Caddy >= 2.9 drop this whole
+        # site, silently disabling the HTTP proxy while the :25 layer4 listener stays up.
+        bind {$ADDRESS:0.0.0.0 ::}

-        # Extra services: LSP, Multiplayer, Debugger (windmill_extra gateway)
-        reverse_proxy /ws/* /ws_mp/* /ws_debug/* http://windmill_extra:3000
+        # Extra services: LSP, Multiplayer, Debugger (windmill_extra gateway).
+        # reverse_proxy only reads its first argument as a matcher, so listing
+        # several paths inline turns the rest into upstream addresses. The paths
+        # have to go through a named matcher.
+        @extra path /ws/* /ws_mp/* /ws_debug/*
+        reverse_proxy @extra http://windmill_extra:3000

        # Search indexer, Enterprise Edition (windmill_indexer:8002)
        # reverse_proxy /api/srch/* http://windmill_indexer:8002
@@ -1,7 +1,7 @@
-ARG DEBIAN_IMAGE=debian:bookworm-slim
-ARG RUST_IMAGE=rust:1.93-slim-bookworm
+ARG DEBIAN_IMAGE=debian:trixie-slim
+ARG RUST_IMAGE=rust:1.97-slim-trixie

-FROM debian:bookworm-slim AS nsjail
+FROM debian:trixie-slim AS nsjail

 WORKDIR /nsjail

@@ -9,12 +9,12 @@ RUN apt-get -y update \
    && apt-get install -y \
    bison=2:3.8.* \
    flex=2.6.* \
-    g++=4:12.2.* \
-    gcc=4:12.2.* \
-    git=1:2.39.* \
+    g++=4:14.2.* \
+    gcc=4:14.2.* \
+    git=1:2.47.* \
    libprotobuf-dev=3.21.* \
    libnl-route-3-dev=3.7.* \
-    make=4.3-4.1 \
+    make=4.4.* \
    pkg-config=1.8.* \
    protobuf-compiler=3.21.*

@@ -44,7 +44,7 @@ FROM rust_base AS windmill_duckdb_ffi_internal_builder

 WORKDIR /windmill-duckdb-ffi-internal

-RUN apt-get update && apt-get install -y clang=1:14.0-55.* libclang-dev=1:14.0-55.* cmake=3.25.* && \
+RUN apt-get update && apt-get install -y clang=1:19.0* libclang-dev=1:19.0* cmake=3.31.* && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

@@ -98,7 +98,7 @@ ARG features=""

 COPY --from=planner /windmill/recipe.json recipe.json

-RUN apt-get update && apt-get install -y libxml2-dev=2.9.* libxmlsec1-dev=1.2.* libkrb5-dev libsasl2-dev libcurl4-openssl-dev clang=1:14.0-55.* libclang-dev=1:14.0-55.* cmake=3.25.* && \
+RUN apt-get update && apt-get install -y libxml2-dev=2.12.* libxmlsec1-dev=1.2.* libkrb5-dev libsasl2-dev libcurl4-openssl-dev clang=1:19.0* libclang-dev=1:19.0* cmake=3.31.* && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

@@ -135,9 +135,8 @@ FROM ${DEBIAN_IMAGE}

 ARG TARGETPLATFORM
 ARG POWERSHELL_VERSION=7.5.0
-ARG POWERSHELL_DEB_VERSION=7.5.0-1
-ARG KUBECTL_VERSION=1.28.7
-ARG HELM_VERSION=3.14.3
+ARG KUBECTL_VERSION=1.36.2
+ARG HELM_VERSION=3.21.2
 # NOTE: If changing, also change go version in workspace dependencies template at WorkspaceDependenciesEditor.svelte
 ARG GO_VERSION=1.26.0
 ARG APP=/usr/src/app
@@ -163,14 +162,15 @@ ENV PATH /usr/local/bin:/root/.local/bin:/tmp/.local/bin:$PATH


 RUN apt-get update \
-    && apt-get install -y --no-install-recommends netbase tzdata ca-certificates wget curl jq unzip build-essential unixodbc xmlsec1 software-properties-common tini gnupg lsb-release \
+    && apt-get upgrade -y \
+    && apt-get install -y --no-install-recommends netbase tzdata ca-certificates wget curl jq unzip build-essential unixodbc xmlsec1 tini gnupg libargon2-1 \
    && if echo "$features" | grep -q "ee"; then apt-get install -y --no-install-recommends libsasl2-modules-gssapi-mit krb5-user; fi \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

 # Install latest PostgreSQL client (pg_dump) from official PostgreSQL apt repository
 RUN curl -fsSL https://www.postgresql.org/media/keys/ACCC4CF8.asc | gpg --dearmor -o /usr/share/keyrings/postgresql-archive-keyring.gpg \
-    && echo "deb [signed-by=/usr/share/keyrings/postgresql-archive-keyring.gpg] https://apt.postgresql.org/pub/repos/apt $(lsb_release -cs)-pgdg main" > /etc/apt/sources.list.d/pgdg.list \
+    && echo "deb [signed-by=/usr/share/keyrings/postgresql-archive-keyring.gpg] https://apt.postgresql.org/pub/repos/apt $(. /etc/os-release; echo "$VERSION_CODENAME")-pgdg main" > /etc/apt/sources.list.d/pgdg.list \
    && apt-get update \
    && apt-get install -y --no-install-recommends postgresql-client \
    && apt-get clean \
@@ -183,12 +183,14 @@ RUN if [ "$WITH_GIT" = "true" ]; then \
    && rm -rf /var/lib/apt/lists/*; \
    else echo 'Building the image without git'; fi;

+# PowerShell ships as a tarball: the upstream .deb depends on libicu<=74 which no longer exists in trixie
 RUN if [ "$WITH_POWERSHELL" = "true" ]; then \
-    if [ "$TARGETPLATFORM" = "linux/amd64" ]; then apt-get update -y && apt install libicu-dev -y && wget -O 'pwsh.deb' "https://github.com/PowerShell/PowerShell/releases/download/v${POWERSHELL_VERSION}/powershell_${POWERSHELL_DEB_VERSION}.deb_amd64.deb" && apt-get clean \
-    && rm -rf /var/lib/apt/lists/* && \
-    dpkg --install 'pwsh.deb' && \
-    rm 'pwsh.deb'; \
-    elif [ "$TARGETPLATFORM" = "linux/arm64" ]; then apt-get update -y && apt install libicu-dev -y && wget -O powershell.tar.gz "https://github.com/PowerShell/PowerShell/releases/download/v${POWERSHELL_VERSION}/powershell-${POWERSHELL_VERSION}-linux-arm64.tar.gz" && apt-get clean \
+    case "$TARGETPLATFORM" in \
+    "linux/amd64") pwsh_arch=x64 ;; \
+    "linux/arm64") pwsh_arch=arm64 ;; \
+    *) pwsh_arch="" ;; \
+    esac; \
+    if [ -n "$pwsh_arch" ]; then apt-get update -y && apt install libicu76 -y && wget -O powershell.tar.gz "https://github.com/PowerShell/PowerShell/releases/download/v${POWERSHELL_VERSION}/powershell-${POWERSHELL_VERSION}-linux-${pwsh_arch}.tar.gz" && apt-get clean \
    && rm -rf /var/lib/apt/lists/* && \
    mkdir -p /opt/microsoft/powershell/7 && \
    tar zxf powershell.tar.gz -C /opt/microsoft/powershell/7 && \
@@ -233,7 +235,7 @@ ENV PATH="${PATH}:/usr/local/go/bin"
 ENV GO_PATH=/usr/local/go/bin/go

 # Install UV
-RUN curl --proto '=https' --tlsv1.2 -LsSf https://github.com/astral-sh/uv/releases/download/0.9.25/uv-installer.sh | sh && mv /root/.local/bin/uv /usr/local/bin/uv
+RUN curl --proto '=https' --tlsv1.2 -LsSf https://github.com/astral-sh/uv/releases/download/0.11.24/uv-installer.sh | sh && mv /root/.local/bin/uv /usr/local/bin/uv

 # Preinstall python runtimes to temp build location (will copy with world-writable perms later)
 # --compile-bytecode precompiles the stdlib to .pyc so jobs don't recompile it on every run
@@ -292,7 +294,7 @@ RUN bun install -g windmill-cli \
 RUN curl -fsSL https://claude.ai/install.sh | bash \
    && cp /root/.local/share/claude/versions/* /usr/bin/claude

-COPY --from=php:8.3.30-cli-bookworm /usr/local/bin/php /usr/bin/php
+COPY --from=php:8.3.30-cli-trixie /usr/local/bin/php /usr/bin/php
 COPY --from=composer:2.9.5 /usr/bin/composer /usr/bin/composer

 # add the docker client to call docker from a worker if enabled
@@ -303,13 +305,13 @@ ENV CARGO_HOME="/tmp/windmill/cache/cargo"
 ENV LD_LIBRARY_PATH="."

 # nsjail runtime deps and binary
-RUN apt-get update && apt-get install -y --no-install-recommends libprotobuf32 libnl-route-3-200 libnl-3-200 \
+RUN apt-get update && apt-get install -y --no-install-recommends libprotobuf32t64 libnl-route-3-200 libnl-3-200 \
    && apt-get clean && rm -rf /var/lib/apt/lists/*
 COPY --from=nsjail /nsjail/nsjail /bin/nsjail

 # crane: pulls + flattens images for the sandboxed container runtime (`# sandbox <image>`).
 # Single static binary — no daemon/store/root needed. See docs/docker-v2-runtime.md.
-ARG CRANE_VERSION=v0.20.6
+ARG CRANE_VERSION=v0.21.7
 RUN arch="$(dpkg --print-architecture)"; \
    case "$arch" in amd64) crane_arch=x86_64 ;; arm64) crane_arch=arm64 ;; *) echo >&2 "error: unsupported arch '$arch' for crane"; exit 1 ;; esac; \
    wget -O /tmp/crane.tgz "https://github.com/google/go-containerregistry/releases/download/${CRANE_VERSION}/go-containerregistry_Linux_${crane_arch}.tar.gz" \
@@ -1,208 +1,14 @@
-# AI Evals Authoring Guide
+# AI Evals

-This folder contains black-box benchmark cases for:
+Black-box benchmark cases for the Windmill AI generation modes (`flow`, `app`,
+`script`, `cli`, `global`).

- `flow`
- `app`
- `script`
- `cli`
- `global`
+**Authoring and running cases is documented in the `ai-evals` skill** — load it
+before adding/changing a case or running a benchmark. Claude Code reads
+`.claude/skills/ai-evals/SKILL.md`; Codex and Pi read
+`.agents/skills/ai-evals/SKILL.md` (same canonical file). Invoke with `/ai-evals` in
+Claude Code, `$ai-evals` in Codex, or `pi --skill ai-evals`.

-The goal is to test the current production prompts and guidance with realistic user requests, not to test one exact implementation shape.
+For AI chat / copilot changes that these evals measure, see the `ai-chat` skill.

-## Core rules
-
-1. Write prompts like a real user request.
-2. Prefer behavior, inputs, constraints, and outcomes over internal implementation details.
-3. Keep deterministic validation narrow and hard.
-4. Put semantic expectations in `judgeChecklist`.
-5. Use `expected` fixtures only when exact structure really matters.
-
-## Prompt writing
-
-Prompts should sound like something a user would naturally ask.
-
-Good:
-
- "Create a flow that routes support requests based on customer tier."
- "Add a reset button that sets the counter back to 0."
- "Create a flow that reuses the existing greeting script instead of duplicating the logic."
-
-Bad:
-
- "Use `branchone` with 3 branches and a default branch."
- "Create a `rawscript` step with this exact topology."
- "This is a benchmark harness."
-
-Do not write prompts as if the user knows Windmill internals unless the case is explicitly testing a power-user workflow.
-
-## Flow-specific rules
-
-This is the main principle you asked for:
-
- flow prompts should read like requests from a user who does not know the product internals
- the user should ask for behavior, not for `branchone`, `branchall`, `rawscript`, `preprocessor_module`, `failure_module`, exact graph topology, or other internal constructs
-
-That means:
-
- creation cases should describe the business behavior and expected result
- modification cases may mention existing step names, because the user can see the current flow
- only mention special Windmill constructs when the case is explicitly about those constructs
-
-Examples:
-
- acceptable creation prompt:
-  "Create a purchase approval flow that pauses for approval and asks the approver for a comment."
- avoid:
-  "Create a suspend step with one required event and a resume form."
-
-For flow cases, do not fail a case just because the model chose a different valid topology.
-
-## App-specific rules
-
-App prompts should focus on user-visible behavior:
-
- what the UI should let the user do
- what should persist
- what backend behavior is needed
-
-Avoid prompting in terms of React structure, component names, or implementation unless the case is specifically about editing an existing app.
-
-## CLI-specific rules
-
-CLI prompts can be more explicit about paths and file names because real CLI users often do specify them.
-
-Still, avoid benchmark phrasing. The prompt should read like a repo task, not a harness instruction.
-
-When relevant, ask the assistant to tell the user which `wmill` commands to run next. That is part of the benchmarked behavior.
-
-## Global-specific rules
-
-Global prompts should exercise workspace-level drafting behavior:
-
- inspecting existing scripts, flows, apps, schedules, triggers, resources, and variables when relevant
- writing AI drafts rather than saving or deploying by default
- producing coherent multi-artifact changes when the request crosses artifact boundaries
-
-Keep deterministic validation focused on the draft contract: required draft type/path, required content snippets, forbidden draft paths, and forbidden mutating tools such as deploy/delete unless the case explicitly asks for them.
-
-Datatable cases should set `skipJudge: true` and validate through tool-use
-(`requiredToolsUsed` / `forbiddenToolsUsed`) and SQL-argument assertions
-(`toolCallArgs` with `stringIncludesAnyOf`, e.g. `['select']`, `['create table']`,
-`['update', 'insert into']`). Two reasons the judge is unreliable here:
-
- `list_datatables`, `get_datatable_table_schema`, and `exec_datatable_sql`
-  produce no drafts, and the global judge only sees the drafts artifact — it
-  scores a no-draft conversational answer as empty (same as the
-  `askUserQuestion` cases).
- Even a case that *does* produce a draft (a script reading the data table via
-  `wmill.datatable()` at runtime) is mis-judged: the judge has no datatable SDK
-  reference and penalizes correct `wmill.datatable()` usage as wrong. Verify the
-  SDK call deterministically instead — `requiredDrafts.valueIncludes: ['wmill.datatable(']`
-  plus forbidding `exec_datatable_sql` (keeping chat-time SQL distinct from
-  runtime SDK use).
-
-`stringIncludesAnyOf` is existential over calls (at least one matching call), so a
-mutation case still passes when the model mixes its UPDATE/INSERT with
-verification SELECTs. The in-memory engine (`datatableSqlEngine.ts`) is stateful
-within a case — writes persist, so a model that re-queries to verify its
-CREATE/UPDATE sees the change and does not loop. But the engine is best-effort
-(SELECT returns all rows of the referenced/first table with no WHERE/projection),
-so still never assert specific returned row values. Seed data via
-`workspace.datatables` in the `initial` fixture (see README).
-
-## Deterministic validation
-
-Use deterministic validation only for hard failures such as:
-
- missing required files
- unexpected extra files when the prompt says not to create them
- syntax errors
- unresolved flow refs
- missing required special modules or suspend config
- obvious artifact corruption
-
-Do not use deterministic validation to enforce one preferred implementation for broad creation tasks.
-
-Examples of bad hard checks:
-
- exact step topology for a creation flow
- exact branch structure when the prompt only asked for routing behavior
- exact input shape when multiple reasonable shapes are acceptable
-
-## Judge checklist
-
-Every non-trivial case should have a `judgeChecklist`.
-
-The checklist should capture:
-
- the user-visible behavior that must be present
- important constraints
- key completion criteria
-
-The checklist should not duplicate low-level implementation details unless they are truly required by the task.
-
-Good checklist items:
-
- "the flow calculates the order total with 8% tax"
- "the app persists recipes appropriately for a raw Windmill app"
- "the flow reuses the existing workspace script instead of rewriting the logic"
-
-Bad checklist items:
-
- "uses `branchone`"
- "contains a `rawscript` node"
-
-## When to use `expected`
-
-Use `expected` fixtures when the case is structure-sensitive, for example:
-
- exact file creation
- exact script content
- modification cases where a specific file must change in a specific way
- cases where preserving an existing structure is part of the requirement
-
-Do not use a full `expected` artifact as the semantic oracle for broad creation tasks when multiple valid outputs should pass.
-
-## When to use `initial`
-
-Use `initial` when the benchmark is about:
-
- editing an existing artifact
- reusing existing workspace assets
- preserving existing behavior while adding a change
-
-If the case is greenfield, prefer no `initial`.
-
-## Case design ladder
-
-Prefer suites that get gradually harder:
-
-1. trivial create case
-2. realistic create case
-3. reuse-existing-assets case
-4. modification case
-5. refactor case
-6. edge-case or niche product behavior
-
-The last cases in a suite should cover unusual or product-specific behavior.
-
-## Anti-patterns
-
-Avoid these:
-
- benchmark framing in prompts
- over-specified internal topology for creation tasks
- judge checklists that just restate implementation details
- deterministic validation that encodes one preferred solution
- fixtures that are so minimal or brittle that they create false negatives
-
-## Before adding a case
-
-Ask:
-
-1. Would a real user plausibly write this prompt?
-2. If the model solves it in a different valid way, would the case still pass?
-3. Are the hard deterministic checks only catching objectively broken output?
-4. Does the `judgeChecklist` describe the real success criteria?
-5. If this case fails, will the reason be understandable from the saved artifacts?
+The full case format, fields, and fixture details remain in `ai_evals/README.md`.
@@ -3,7 +3,7 @@ import { tmpdir } from "os";
 import { join } from "path";
 import type { AIProvider } from "$lib/gen/types.gen";
 import {
-  globalTools,
+  globalToolsFor,
  prepareGlobalSystemMessage,
  prepareGlobalUserMessage,
 } from "../../../../../frontend/src/lib/components/copilot/chat/global/core";
@@ -43,6 +43,67 @@ const LIVE_EDITOR_ITEM_KINDS = {
  app: "raw_app",
 } as const;

+// SessionArtifactsStore can't run here (bun has no IndexedDB, nor the compiled $state runes),
+// so mirror only its tool-facing shape; its own logic (scoping, race guard) is unit-tested.
+const EVAL_SESSION_ID = "eval-session";
+function createEvalArtifactHelpers() {
+  const items = new Map<string, Record<string, unknown>>();
+  let seq = 0;
+  const store = {
+    create: async (sessionId: string, input: Record<string, any>) => {
+      const now = seq++;
+      const artifact = {
+        id: `eval-artifact-${now}`,
+        sessionId,
+        chatId: input.chatId,
+        kind: input.kind ?? "md",
+        name: input.name,
+        content: input.content,
+        createdAt: now,
+        updatedAt: now,
+      };
+      items.set(artifact.id, artifact);
+      return artifact;
+    },
+    get: async (id: string) => items.get(id),
+    update: async (
+      id: string,
+      input: Record<string, any>,
+      opts?: { sessionId?: string },
+    ) => {
+      const existing = items.get(id);
+      if (!existing) return undefined;
+      if (
+        opts?.sessionId !== undefined &&
+        existing.sessionId !== opts.sessionId
+      )
+        return undefined;
+      const updated = {
+        ...existing,
+        name: input.name ?? existing.name,
+        content: input.content ?? existing.content,
+        updatedAt: seq++,
+      };
+      items.set(id, updated);
+      return updated;
+    },
+    remove: async (id: string) => {
+      items.delete(id);
+    },
+    listForSession: async (sessionId: string) =>
+      [...items.values()].filter((a) => a.sessionId === sessionId),
+  };
+  return {
+    helpers: {
+      artifacts: store,
+      sessionId: EVAL_SESSION_ID,
+      getChatId: () => "eval-chat",
+      openArtifact: () => {},
+    },
+    snapshot: () => [...items.values()],
+  };
+}
+
 export interface GlobalLiveEditorDraftFixture {
  type: keyof typeof LIVE_EDITOR_ITEM_KINDS;
  storagePath?: string;
@@ -80,6 +141,8 @@ export interface GlobalEvalOptions {
  workspaceFixtures?: BenchmarkWorkspaceRunnables;
  liveEditorDrafts?: GlobalLiveEditorDraftFixture[];
  user?: GlobalUserFixture;
+  // Emulate a session chat (preview tools + session prompt); default false = standalone baseline.
+  sessionChat?: boolean;
  model?: string;
  maxIterations?: number;
  provider?: AIProvider;
@@ -98,7 +161,10 @@ export async function runGlobalEval(
    (await mkdtemp(join(tmpdir(), "wmill-frontend-global-benchmark-")));

  clearGlobalDrafts(workspaceRoot);
-  registerBenchmarkWorkspaceRunnables(workspaceRoot, options.workspaceFixtures ?? {});
+  registerBenchmarkWorkspaceRunnables(
+    workspaceRoot,
+    options.workspaceFixtures ?? {},
+  );
  seedLiveEditorDrafts(workspaceRoot, options.liveEditorDrafts ?? []);

  try {
@@ -107,18 +173,25 @@ export async function runGlobalEval(
      process.env[DISABLE_ACTIVE_EDITOR_CONTEXT_ENV] !== "1";
    // Pass the seeded identity straight to the prompt builder rather than mutating
    // the process-global `userStore`, so concurrent cases never race on it.
+    const evalArtifacts = createEvalArtifactHelpers();
    const rawResult = await runEval({
      userPrompt,
-      systemMessage: prepareGlobalSystemMessage(undefined, { user: options.user }),
+      systemMessage: prepareGlobalSystemMessage(undefined, {
+        user: options.user,
+        previewTools: options.sessionChat ?? false,
+      }),
      userMessage: prepareGlobalUserMessage(
        userPrompt,
        [],
        injectActiveEditorContext ? { workspace: workspaceRoot } : {},
      ),
-      tools: getGlobalEvalTools(),
-      helpers: {},
+      tools: getGlobalEvalTools(options.sessionChat ?? false),
+      helpers: evalArtifacts.helpers,
      apiKey,
-      getOutput: () => collectGlobalDraftState(workspaceRoot),
+      getOutput: async () => ({
+        ...(await collectGlobalDraftState(workspaceRoot)),
+        artifacts: evalArtifacts.snapshot(),
+      }),
      onAssistantMessageStart: options.runContext?.onAssistantMessageStart,
      onAssistantToken: options.runContext?.onAssistantChunk,
      onAssistantMessageEnd: options.runContext?.onAssistantMessageEnd,
@@ -213,10 +286,15 @@ function clearLiveEditorDrafts(
  }
 }

-function getGlobalEvalTools(): ProductionTool<{}>[] {
+// Gate session-preview tools on sessionChat, as production's globalToolsFor does.
+function getGlobalEvalTools(sessionChat: boolean): ProductionTool<{}>[] {
  const disableSearchApp = process.env[DISABLE_SEARCH_APP_ENV] === "1";
-  return (globalTools as ProductionTool<{}>[])
-    .filter((tool) => !(disableSearchApp && tool.def.function.name === "search_app"))
+  return (
+    globalToolsFor({ sessionPreview: sessionChat }) as ProductionTool<{}>[]
+  )
+    .filter(
+      (tool) => !(disableSearchApp && tool.def.function.name === "search_app"),
+    )
    .map((tool) => {
      if (!MUTATING_GLOBAL_TOOLS.has(tool.def.function.name)) {
        return tool;
@@ -10,7 +10,9 @@ import type {
 import type {
 	DataTableTables,
 	DataTableTableSchema,
+	EndpointTool,
 	GetDraftForUserResponse,
+	GetOwnDraftResponse,
 	ListDraftsResponse,
 	ScriptLang,
 	UpdateDraftResponse,
@@ -90,6 +92,13 @@ export function resetBenchmarkMockBackend(): void {
 	benchmarkDrafts.clear()
 }

+// Stand-in for FolderService.createFolder so the global create_folder tool runs in
+// memory instead of mutating the real backend. Folders aren't otherwise modelled
+// (no folder-listing in evals), so this just echoes the created name.
+export function createBenchmarkFolder(_workspace: string, name: string): string {
+	return name
+}
+
 export function registerBenchmarkWorkspace(workspace: string): void {
 	benchmarkWorkspaces.add(workspace)
 }
@@ -287,21 +296,28 @@ export function getBenchmarkJobLogs(workspace: string, jobId: string): string {
 /**
 * In-memory stand-in for the per-user draft backend (`DraftService`). The global
 * AI chat now persists and reads drafts through the backend DB instead of an
- * in-tab `UserDraft` cell, so the eval mocks the three draft endpoints it
- * exercises (`updateDraft` / `getDraftForUser` / `listDrafts`) and keeps the
+ * in-tab `UserDraft` cell, so the eval mocks the draft endpoints it exercises
+ * (`updateDraft` / `getOwnDraft` / `getDraftForUser` / `listDrafts`) and keeps the
 * saved values here, keyed by workspace + draft kind + storage path. Mirrors the
 * semantics of the production unit test's mock in
 * `frontend/src/lib/components/copilot/chat/global/core.test.ts`.
 */
 const benchmarkDrafts = new Map<
 	string,
-	{ workspace: string; kind: UserDraftItemKind; path: string; value: unknown }
+	{ workspace: string; kind: UserDraftItemKind; path: string; value: unknown; createdAt: string }
 >()

-// Fixed timestamp so artifacts stay deterministic. No eval simulates a
-// concurrent writer, so every save is accepted and the conflict branch is
-// never taken — the syncer just records this as its `last_sync` baseline.
-const BENCHMARK_DRAFT_TIMESTAMP = '1970-01-01T00:00:00.000Z'
+// Counter-based timestamps: deterministic run-to-run (same event order → same
+// values) but MONOTONIC per update, because production bumps a draft row's
+// created_at on every upsert and the diff snapshot cache keys patch reuse on
+// it — a fixed timestamp would serve stale patches after an edit. No eval
+// simulates a concurrent writer, so every save is accepted and the conflict
+// branch is never taken.
+let benchmarkDraftClock = 0
+function nextBenchmarkDraftTimestamp(): string {
+	benchmarkDraftClock += 1
+	return new Date(benchmarkDraftClock * 1000).toISOString()
+}

 function benchmarkDraftKey(workspace: string, kind: string, path: string): string {
 	return `${workspace}::${kind}::${path}`
@@ -332,7 +348,8 @@ export function seedBenchmarkDraft(
 		workspace,
 		kind,
 		path,
-		value
+		value,
+		createdAt: nextBenchmarkDraftTimestamp()
 	})
 }

@@ -345,6 +362,7 @@ export function updateBenchmarkDraft(input: {
 }): UpdateDraftResponse {
 	const key = benchmarkDraftKey(input.workspace, input.kind, input.path)
 	const value = input.requestBody?.value
+	const createdAt = nextBenchmarkDraftTimestamp()
 	if (value == null) {
 		benchmarkDrafts.delete(key)
 	} else {
@@ -352,10 +370,11 @@ export function updateBenchmarkDraft(input: {
 			workspace: input.workspace,
 			kind: input.kind,
 			path: input.path,
-			value
+			value,
+			createdAt
 		})
 	}
-	return { status: 'saved', current_timestamp: BENCHMARK_DRAFT_TIMESTAMP }
+	return { status: 'saved', current_timestamp: createdAt }
 }

 /** Mirror `DraftService.getDraftForUser`: 404-shaped throw when absent so the
@@ -369,7 +388,32 @@ export function getBenchmarkDraftForUser(input: {
 	if (!entry) {
 		throw Object.assign(new Error(`no draft for "${input.path}"`), { status: 404 })
 	}
-	return { value: entry.value, created_at: BENCHMARK_DRAFT_TIMESTAMP }
+	return { value: entry.value, created_at: entry.createdAt }
+}
+
+/** Mirror `DraftService.getOwnDraft`: `null` (200) when absent — unlike
+ * `getDraftForUser`, absence is not an error on this route. */
+export function getBenchmarkOwnDraft(input: {
+	workspace: string
+	kind: UserDraftItemKind
+	path: string
+}): GetOwnDraftResponse {
+	const entry = benchmarkDrafts.get(benchmarkDraftKey(input.workspace, input.kind, input.path))
+	if (!entry) {
+		return null
+	}
+	return { value: entry.value, created_at: entry.createdAt }
+}
+
+/** Whether a deployed benchmark item exists for a draft row's kind+path —
+ * drives `draft_only`, which production computes against the deployed tables. */
+function benchmarkDeployedExists(workspace: string, kind: UserDraftItemKind, path: string): boolean {
+	if (kind === 'script') return Boolean(getBenchmarkScriptByPath(workspace, path))
+	if (kind === 'flow') return Boolean(getBenchmarkFlowByPath(workspace, path))
+	if (kind === 'app' || kind === 'raw_app') return Boolean(getBenchmarkAppByPath(workspace, path))
+	// Drawer kinds (variables/resources/schedules/triggers) have no deployed
+	// benchmark stores today.
+	return false
 }

 /** Mirror `DraftService.listDrafts`: metadata rows (no value) for a workspace. */
@@ -380,9 +424,9 @@ export function listBenchmarkDrafts(workspace: string): ListDraftsResponse {
 			kind: entry.kind,
 			path: entry.path,
 			summary: (entry.value as { summary?: string } | null)?.summary,
-			draft_only: true,
+			draft_only: !benchmarkDeployedExists(workspace, entry.kind, entry.path),
 			legacy_draft: false,
-			created_at: BENCHMARK_DRAFT_TIMESTAMP
+			created_at: entry.createdAt
 		}))
 }

@@ -672,3 +716,138 @@ function buildBenchmarkApp(app: BenchmarkWorkspaceApp): AppWithLastVersion {
 		raw_app: true
 	}
 }
+
+// ============= API endpoint catalog (McpService.listMcpTools + raw fetch) =============
+// The global chat's API catalog tools list endpoints via McpService and execute
+// them with a plain relative fetch('/api/...'), which has no meaning in the
+// vitest environment. A representative slice of the real catalog is served here,
+// and `handleBenchmarkApiFetch` answers the executed calls.
+
+const BENCHMARK_MCP_TOOLS: EndpointTool[] = [
+	{
+		name: 'listWorkers',
+		description: 'List workers',
+		instructions: 'List all workers with their last ping and job counts.',
+		path: '/workers/list',
+		method: 'GET',
+		query_params_schema: {
+			type: 'object',
+			properties: { page: { type: 'integer' }, per_page: { type: 'integer' } }
+		}
+	},
+	{
+		name: 'listQueue',
+		description: 'List queued jobs',
+		instructions: '',
+		path: '/w/{workspace}/jobs/queue/list',
+		method: 'GET',
+		path_params_schema: {
+			type: 'object',
+			properties: { workspace: { type: 'string' } },
+			required: ['workspace']
+		}
+	},
+	{
+		name: 'runScriptByPath',
+		description: 'Run the deployed version of a script by path',
+		instructions: '',
+		path: '/w/{workspace}/jobs/run/p/{path}',
+		method: 'POST',
+		path_params_schema: {
+			type: 'object',
+			properties: { workspace: { type: 'string' }, path: { type: 'string' } },
+			required: ['workspace', 'path']
+		},
+		body_schema: { type: 'object', properties: {} }
+	},
+	{
+		name: 'runFlowByPath',
+		description: 'Run the deployed version of a flow by path',
+		instructions: '',
+		path: '/w/{workspace}/jobs/run/f/{path}',
+		method: 'POST',
+		path_params_schema: {
+			type: 'object',
+			properties: { workspace: { type: 'string' }, path: { type: 'string' } },
+			required: ['workspace', 'path']
+		},
+		body_schema: { type: 'object', properties: {} }
+	},
+	// Draft-covered endpoints, present so steering cases exercise the guard the
+	// way production does (hidden from search, refused at call time).
+	{
+		name: 'getScriptByPath',
+		description: 'Get a script by path',
+		instructions: '',
+		path: '/w/{workspace}/scripts/get/p/{path}',
+		method: 'GET'
+	},
+	{
+		name: 'createFlow',
+		description: 'Create a flow',
+		instructions: '',
+		path: '/w/{workspace}/flows/create',
+		method: 'POST'
+	},
+	{
+		name: 'deleteSchedule',
+		description: 'Delete a schedule',
+		instructions: '',
+		path: '/w/{workspace}/schedules/delete/{path}',
+		method: 'DELETE'
+	},
+	{
+		name: 'getVariable',
+		description: 'Get a variable',
+		instructions: '',
+		path: '/w/{workspace}/variables/get/{path}',
+		method: 'GET'
+	}
+]
+
+export function listBenchmarkMcpTools(): EndpointTool[] {
+	return BENCHMARK_MCP_TOOLS
+}
+
+const BENCHMARK_WORKERS = [
+	{
+		worker: 'wk-benchmark-1',
+		worker_instance: 'benchmark-host',
+		last_ping: 2,
+		started_at: BENCHMARK_TIMESTAMP,
+		jobs_executed: 42,
+		custom_tags: null,
+		worker_group: 'default',
+		wm_version: 'benchmark'
+	},
+	{
+		worker: 'wk-benchmark-2',
+		worker_instance: 'benchmark-host',
+		last_ping: 5,
+		started_at: BENCHMARK_TIMESTAMP,
+		jobs_executed: 17,
+		custom_tags: null,
+		worker_group: 'default',
+		wm_version: 'benchmark'
+	}
+]
+
+/** True when `handleBenchmarkApiFetch` has an answer for this `/api/...` url.
+ * Any other relative fetch must keep its normal (non-benchmark) behavior —
+ * intercepting it with a synthetic 404 sends the model into retry loops. */
+export function hasBenchmarkApiHandler(url: string): boolean {
+	const path = url.split('?')[0]
+	return path === '/api/workers/list' || /^\/api\/w\/[^/]+\/jobs\/queue\/list$/.test(path)
+}
+
+/** Answer a relative `/api/...` fetch issued by the API catalog executor. */
+export function handleBenchmarkApiFetch(url: string): Response {
+	const path = url.split('?')[0]
+	if (path === '/api/workers/list') {
+		return Response.json(BENCHMARK_WORKERS)
+	}
+	if (/^\/api\/w\/[^/]+\/jobs\/queue\/list$/.test(path)) {
+		return Response.json([])
+	}
+	return Response.json({ error: `no benchmark handler for ${path}` }, { status: 404 })
+}
@@ -2,6 +2,7 @@ import { afterEach, beforeEach, describe, expect, it } from 'bun:test'
 import {
 	clearBenchmarkDrafts,
 	getBenchmarkDraftForUser,
+	getBenchmarkOwnDraft,
 	listBenchmarkDrafts,
 	resetBenchmarkMockBackend,
 	seedBenchmarkDraft,
@@ -55,6 +56,27 @@ describe('mockBackend drafts', () => {
 		expect(() => getBenchmarkDraftForUser({ workspace: WORKSPACE, kind: 'variable', path: 'f/evals/token' })).toThrow()
 	})

+	it('returns null from getOwnDraft when no draft exists', () => {
+		expect(
+			getBenchmarkOwnDraft({ workspace: WORKSPACE, kind: 'trigger_schedule', path: 'u/evals/missing' })
+		).toBeNull()
+	})
+
+	// The global chat hydrates drawer-kind drafts (schedule/trigger/resource/variable)
+	// through getOwnDraft — getDraftForUser rejects those kinds as private.
+	it('hydrates a saved drawer-kind draft through getOwnDraft', () => {
+		const value = { path: 'u/evals/nightly', schedule: '0 0 9 * * *' }
+		updateBenchmarkDraft({
+			workspace: WORKSPACE,
+			kind: 'trigger_schedule',
+			path: 'u/evals/nightly',
+			requestBody: { value }
+		})
+		expect(
+			getBenchmarkOwnDraft({ workspace: WORKSPACE, kind: 'trigger_schedule', path: 'u/evals/nightly' })?.value
+		).toEqual(value)
+	})
+
 	it('throws a 404-shaped error when no draft exists', () => {
 		try {
 			getBenchmarkDraftForUser({ workspace: WORKSPACE, kind: 'script', path: 'f/evals/missing' })
@@ -3,6 +3,20 @@ import { expect, it, vi } from 'vitest'
 import { mkdir, writeFile } from 'fs/promises'
 // @ts-ignore - Node.js path
 import { dirname, resolve } from 'path'
+import { handleBenchmarkApiFetch, hasBenchmarkApiHandler } from './mockBackend'
+
+// The API catalog executor issues relative fetch('/api/...') calls, which have
+// no meaning in the vitest environment — serve the ones the benchmark handles.
+// Every other relative fetch keeps its normal behavior (it fails the same way
+// it does without this stub) so unrelated tools see an unchanged environment.
+const ORIGINAL_FETCH = globalThis.fetch
+globalThis.fetch = (async (input: unknown, init?: RequestInit) => {
+	const url = typeof input === 'string' ? input : ((input as Request | URL | null)?.url ?? '')
+	if (typeof url === 'string' && hasBenchmarkApiHandler(url)) {
+		return handleBenchmarkApiFetch(url)
+	}
+	return ORIGINAL_FETCH(input as Parameters<typeof fetch>[0], init)
+}) as typeof fetch

 vi.mock('monaco-editor', () => ({
 	editor: {},
@@ -40,6 +54,7 @@ vi.mock('$lib/gen', async () => {
 		getBenchmarkDraftForUser,
 		getBenchmarkFlowByPath,
 		getBenchmarkJobLogs,
+		getBenchmarkOwnDraft,
 		getBenchmarkScriptByHash,
 		getBenchmarkScriptByPath,
 		hasBenchmarkWorkspace,
@@ -49,13 +64,15 @@ vi.mock('$lib/gen', async () => {
 		listBenchmarkFlows,
 		listBenchmarkJobs,
 		listBenchmarkScripts,
+		createBenchmarkFolder,
 		createBenchmarkHttpTrigger,
 		createBenchmarkSchedule,
 		previewBenchmarkSchedule,
 		runBenchmarkDatatableSql,
 		runBenchmarkFlowByPath,
 		runBenchmarkScriptPreview,
-		updateBenchmarkDraft
+		updateBenchmarkDraft,
+		listBenchmarkMcpTools
 	} = await import('./mockBackend')

 	function wrapService<T extends object>(target: T, overrides: Record<string, unknown>): T {
@@ -85,11 +102,21 @@ vi.mock('$lib/gen', async () => {
 				hasBenchmarkWorkspace(data.workspace)
 					? getBenchmarkDraftForUser(data)
 					: actual.DraftService.getDraftForUser(data),
+			getOwnDraft: async (data: { workspace: string; kind: any; path: string }) =>
+				hasBenchmarkWorkspace(data.workspace)
+					? getBenchmarkOwnDraft(data)
+					: actual.DraftService.getOwnDraft(data),
 			listDrafts: async (data: { workspace: string }) =>
 				hasBenchmarkWorkspace(data.workspace)
 					? listBenchmarkDrafts(data.workspace)
 					: actual.DraftService.listDrafts(data)
 		}),
+		FolderService: wrapService(actual.FolderService, {
+			createFolder: async (data: { workspace: string; requestBody: { name: string } }) =>
+				hasBenchmarkWorkspace(data.workspace)
+					? createBenchmarkFolder(data.workspace, data.requestBody.name)
+					: actual.FolderService.createFolder(data)
+		}),
 		ScriptService: wrapService(actual.ScriptService, {
 			listScripts: async (data: { workspace: string }) =>
 				hasBenchmarkWorkspace(data.workspace)
@@ -99,13 +126,32 @@ vi.mock('$lib/gen', async () => {
 				hasBenchmarkWorkspace(data.workspace)
 					? Boolean(getBenchmarkScriptByPath(data.workspace, data.path))
 					: actual.ScriptService.existsScriptByPath(data),
-			getScriptByPath: async (data: { workspace: string; path: string }) => {
+			getScriptByPath: async (data: { workspace: string; path: string; getDraft?: boolean }) => {
 				if (hasBenchmarkWorkspace(data.workspace)) {
 					const script = getBenchmarkScriptByPath(data.workspace, data.path)
+					// `getDraft` mirrors production's overlay: the row plus the caller's
+					// draft and a `no_deployed` marker (draft-only item). The diff tool
+					// reads through this shape — without it every draft looks absent.
+					const draft = data.getDraft
+						? getBenchmarkOwnDraft({ workspace: data.workspace, kind: 'script', path: data.path })
+						: null
 					if (!script) {
-						throw new Error(`Script "${data.path}" not found in benchmark workspace`)
+						if (data.getDraft && draft) {
+							return {
+								...(draft.value as Record<string, unknown>),
+								path: data.path,
+								draft: draft.value,
+								no_deployed: true
+							}
+						}
+						throw Object.assign(
+							new Error(`Script "${data.path}" not found in benchmark workspace`),
+							{ status: 404 }
+						)
 					}
-					return script
+					return data.getDraft
+						? { ...script, draft: draft?.value ?? undefined, no_deployed: false }
+						: script
 				}
 				return actual.ScriptService.getScriptByPath(data)
 			},
@@ -139,13 +185,30 @@ vi.mock('$lib/gen', async () => {
 				hasBenchmarkWorkspace(data.workspace)
 					? Boolean(getBenchmarkFlowByPath(data.workspace, data.path))
 					: actual.FlowService.existsFlowByPath(data),
-			getFlowByPath: async (data: { workspace: string; path: string }) => {
+			getFlowByPath: async (data: { workspace: string; path: string; getDraft?: boolean }) => {
 				if (hasBenchmarkWorkspace(data.workspace)) {
 					const flow = getBenchmarkFlowByPath(data.workspace, data.path)
+					// Mirror production's `getDraft` overlay (see getScriptByPath above).
+					const draft = data.getDraft
+						? getBenchmarkOwnDraft({ workspace: data.workspace, kind: 'flow', path: data.path })
+						: null
 					if (!flow) {
-						throw new Error(`Flow "${data.path}" not found in benchmark workspace`)
+						if (data.getDraft && draft) {
+							return {
+								...(draft.value as Record<string, unknown>),
+								path: data.path,
+								draft: draft.value,
+								no_deployed: true
+							}
+						}
+						throw Object.assign(
+							new Error(`Flow "${data.path}" not found in benchmark workspace`),
+							{ status: 404 }
+						)
 					}
-					return flow
+					return data.getDraft
+						? { ...flow, draft: draft?.value ?? undefined, no_deployed: false }
+						: flow
 				}
 				return actual.FlowService.getFlowByPath(data)
 			},
@@ -287,6 +350,12 @@ vi.mock('$lib/gen', async () => {
 			queryResourceTypes: async (data: { workspace: string }) =>
 				hasBenchmarkWorkspace(data.workspace) ? [] : actual.ResourceService.queryResourceTypes(data)
 		}),
+		McpService: wrapService(actual.McpService, {
+			listMcpTools: async (data: { workspace: string }) =>
+				hasBenchmarkWorkspace(data.workspace)
+					? listBenchmarkMcpTools()
+					: actual.McpService.listMcpTools(data)
+		}),
 		VariableService: wrapService(actual.VariableService, {
 			existsVariable: async (data: { workspace: string; path: string }) =>
 				hasBenchmarkWorkspace(data.workspace) ? false : actual.VariableService.existsVariable(data),
@@ -308,13 +377,37 @@ vi.mock('$lib/gen', async () => {
 				hasBenchmarkWorkspace(data.workspace)
 					? (listBenchmarkApps(data.workspace) ?? [])
 					: actual.AppService.listApps(data),
-			getAppByPath: async (data: { workspace: string; path: string }) => {
+			getAppByPath: async (data: {
+				workspace: string
+				path: string
+				getDraft?: boolean
+				rawApp?: boolean
+			}) => {
 				if (hasBenchmarkWorkspace(data.workspace)) {
 					const app = getBenchmarkAppByPath(data.workspace, data.path)
+					// Mirror production's `getDraft` overlay (see getScriptByPath above).
+					// Benchmark app drafts live under the raw_app kind.
+					const draft = data.getDraft
+						? getBenchmarkOwnDraft({ workspace: data.workspace, kind: 'raw_app', path: data.path })
+						: null
 					if (!app) {
-						throw new Error(`App "${data.path}" not found in benchmark workspace`)
+						if (data.getDraft && draft) {
+							return {
+								...(draft.value as Record<string, unknown>),
+								path: data.path,
+								raw_app: true,
+								draft: draft.value,
+								no_deployed: true
+							}
+						}
+						throw Object.assign(
+							new Error(`App "${data.path}" not found in benchmark workspace`),
+							{ status: 404 }
+						)
 					}
-					return app
+					return data.getDraft
+						? { ...app, draft: draft?.value ?? undefined, no_deployed: false }
+						: app
 				}
 				return actual.AppService.getAppByPath(data)
 			}
@@ -359,6 +359,8 @@
      - request_approval
      - finalize_purchase
    topLevelStepTypes:
+      - id: request_approval
+        type: [rawscript, script]
      - id: finalize_purchase
        type: rawscript
    schemaRequiredPaths:
@@ -373,6 +375,7 @@
  judgeChecklist:
    - "the flow includes an approval step named `request_approval`"
    - "`request_approval` pauses the flow and asks the approver for a comment"
+    - "`request_approval` is a real script step that generates approval/resume URLs (e.g. via `getResumeUrls`) so approvers receive an actionable link, not a no-op passthrough (identity) step"
    - one approval is enough to continue
    - "the flow includes a final step named `finalize_purchase`"
    - "`finalize_purchase` returns an approved status object after approval"
@@ -3,6 +3,7 @@
    Create a draft Bun script at `f/evals/global/greet_user`.
    It should take a string `name` input and return `Hello, ${name}!`.
    Leave it as an AI draft only; do not deploy or save it.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json
  runtime:
    maxTurns: 10
  validate:
@@ -871,6 +872,240 @@
    - fetches the logs for the requested job id
    - explains the failure from the returned logs (connection refused to the upstream API)

+# --- Page navigation (open_page) ---
+# The assistant should take the user to a Windmill page (Runs/Schedules) with the
+# right filters via open_page, rather than describing where to click or dumping the
+# data. No draft is produced, so the global judge is skipped and we validate the
+# tool call and its arguments.
+
+- id: global-openpage1-runs-failed-of-script
+  prompt: |-
+    Take me to the failed runs of the script at f/evals/global/greet_user so I can see what's going wrong.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - write_script
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - runs
+      - tool: open_page
+        field: status
+        stringIncludesAnyOf:
+          - failure
+      - tool: open_page
+        field: path
+        stringIncludesAnyOf:
+          - f/evals/global/greet_user
+  skipJudge: true
+  judgeChecklist:
+    - opens the Runs page filtered to the failed runs of f/evals/global/greet_user
+    - applies both the failure status and the script path as filters
+    - does not write, deploy, or delete anything
+
+- id: global-openpage2-runs-of-schedule
+  prompt: |-
+    Open the runs page filtered to the jobs triggered by the schedule f/evals/global/nightly_digest.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - write_script
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - runs
+      - tool: open_page
+        field: schedule_path
+        stringIncludesAnyOf:
+          - f/evals/global/nightly_digest
+  skipJudge: true
+  judgeChecklist:
+    - opens the Runs page filtered to jobs triggered by the f/evals/global/nightly_digest schedule
+    - passes the schedule path as the filter
+    - does not write, deploy, or delete anything
+
+- id: global-openpage3-open-schedule
+  prompt: |-
+    Open the schedule f/evals/global/nightly_digest so I can review and edit it.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - write_schedule
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - schedules
+      - tool: open_page
+        field: open
+        stringIncludesAnyOf:
+          - f/evals/global/nightly_digest
+  skipJudge: true
+  judgeChecklist:
+    - opens the Schedules page and targets the f/evals/global/nightly_digest schedule for editing
+    - passes the schedule path so the editor opens on it
+    - does not write, deploy, or delete anything
+
+- id: global-openpage4-workspace-settings-tab
+  prompt: |-
+    Take me to the Git sync configuration for this workspace.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - workspace_settings
+      - tool: open_page
+        field: tab
+        stringIncludesAnyOf:
+          - git_sync
+  skipJudge: true
+  judgeChecklist:
+    - opens the Workspace settings page on the git_sync tab
+    - does not write, deploy, or delete anything
+
+- id: global-openpage5-audit-logs-user
+  prompt: |-
+    Open the audit logs filtered to actions performed by the user admin.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - audit_logs
+      - tool: open_page
+        field: username
+        stringIncludesAnyOf:
+          - admin
+  skipJudge: true
+  judgeChecklist:
+    - opens the Audit logs page filtered to the admin user
+    - does not write, deploy, or delete anything
+
+- id: global-openpage6-triggers-kind
+  prompt: |-
+    Take me to the Kafka triggers for this workspace.
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - open_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - triggers
+      - tool: open_page
+        field: trigger_kind
+        stringIncludesAnyOf:
+          - kafka
+  skipJudge: true
+  judgeChecklist:
+    - opens the Kafka triggers page
+    - does not write, deploy, or delete anything
+
+- id: global-openpage7-compare-review
+  prompt: |-
+    Create a TypeScript script draft at f/evals/global/compare_review_demo that returns the string "ok" (no need to test it), then open the review page so I can look over the pending change and deploy it myself.
+  runtime:
+    maxTurns: 8
+  validate:
+    draftCountExactly: 1
+  toolExpect:
+    requiredToolsUsed:
+      - write_script
+      - open_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - compare
+      # The eval chat is untracked (no modified-items mask), so the model must scope
+      # the review by passing the item it changed explicitly — an omitted mask would
+      # preselect every pending change in the workspace.
+      - tool: open_page
+        field: items
+        stringIncludesAnyOf:
+          - f/evals/global/compare_review_demo
+  skipJudge: true
+  judgeChecklist:
+    - creates the script draft, then opens the Compare & Deploy review page instead of deploying itself
+    - preselects only the created script on the review page
+    - does not deploy or delete anything
+
+- id: global-closepage1-close-runs-tab
+  prompt: |-
+    You just opened the runs page for me in the side panel. Close that tab, I'm done looking at it.
+  runtime:
+    maxTurns: 6
+    sessionChat: true
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - close_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: close_page
+        field: match
+        stringIncludesAnyOf:
+          - runs
+  skipJudge: true
+  judgeChecklist:
+    - closes the runs preview tab in the side panel
+    - does not write, deploy, or delete anything
+
 # --- Documentation search (search_docs) ---
 # Pure product-knowledge questions: the assistant should consult the docs via
 # search_docs and answer conversationally, not draft or mutate anything. No
@@ -1178,7 +1413,7 @@
    when a new hire joins. Keep it as a draft.
  initial: ai_evals/fixtures/frontend/global/initial/user_admin_folders.json
  runtime:
-    maxTurns: 6
+    maxTurns: 8
  validate:
    draftCountExactly: 0
  toolExpect:
@@ -1221,3 +1456,327 @@
    - drafts a flow that returns the current date as an ISO string
    - places it in team_a (writable by this non-admin user) and not team_b (read-only)
    - leaves the result as a draft only
+
+- id: global-test-pipeline-create-node
+  prompt: |-
+    Set up the first step of a data pipeline at `f/evals/global/orders_ingest`.
+    On a schedule, it should pull raw orders and land them in a managed DuckLake
+    table so later steps can build on it. Keep it as an AI draft only — don't
+    deploy or save it.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json
+  runtime:
+    maxTurns: 10
+  validate:
+    draftCountExactly: 1
+    requiredDrafts:
+      - type: script
+        path: f/evals/global/orders_ingest
+        valueIncludes:
+          - pipeline
+    forbiddenDrafts:
+      - type: flow
+        pathStartsWith: f/evals/global/
+  toolExpect:
+    requiredToolsUsed:
+      - write_script
+    forbiddenToolsUsed:
+      - write_flow
+      - deploy_workspace_item
+      - delete_workspace_item
+  judgeChecklist:
+    - builds a data pipeline node as a script (not a flow)
+    - marks the script as a pipeline member with the pipeline annotation in the script's comment syntax (`-- pipeline` for a DuckDB/SQL node, not `// pipeline`)
+    - declares a schedule trigger and writes its output to a managed DuckLake table
+    - leaves the result as an AI draft and does not deploy or save it
+
+- id: global-test-pipeline-two-node-chain
+  prompt: |-
+    Build a small data pipeline in the `f/evals/global` folder: one step that
+    ingests orders into a DuckLake table, and a second step that reads that table
+    and writes a daily order-count rollup table. Wire the second step to run off
+    the first step's output. Keep everything as drafts — don't deploy.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json
+  runtime:
+    maxTurns: 14
+  validate:
+    draftCountAtLeast: 2
+    forbiddenDrafts:
+      - type: flow
+        pathStartsWith: f/evals/global/
+  toolExpect:
+    requiredToolsUsed:
+      - write_script
+    forbiddenToolsUsed:
+      - write_flow
+      - deploy_workspace_item
+      - delete_workspace_item
+  judgeChecklist:
+    - creates two data pipeline nodes as scripts (not a flow) in f/evals/global
+    - both scripts carry the pipeline annotation in their comment syntax (`-- pipeline` for DuckDB/SQL nodes, not `// pipeline`)
+    - the first ingests orders into a DuckLake table
+    - the second reads that same table and writes a daily rollup, wired to the first step's output asset
+    - leaves both as AI drafts without deploying
+
+- id: global-path5-create-folder-then-draft
+  prompt: |-
+    Create a new shared folder called "analytics" for our data work, then draft a
+    script in it that returns the current timestamp as an ISO string. Keep the
+    script as a draft.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_empty.json
+  runtime:
+    maxTurns: 10
+  validate:
+    draftCountExactly: 1
+    requiredDrafts:
+      - type: script
+        pathStartsWith: f/analytics/
+  toolExpect:
+    requiredToolsUsed:
+      - create_folder
+      - write_script
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+  skipJudge: true
+  judgeChecklist:
+    - creates a new shared folder named "analytics" via create_folder
+    - drafts a script placed in that folder (f/analytics/...) returning an ISO timestamp
+    - leaves the script as a draft only
+
+- id: global-artifact-plan-create
+  prompt: |-
+    I'm about to build a customer onboarding flow, but first I want a short written plan I can review and iterate on before any code.
+    Draft a markdown plan with a title, a one-sentence summary, and three or four bullet steps.
+    Keep it as something I can reopen and revise later — don't build the flow itself yet.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_empty.json
+  runtime:
+    maxTurns: 6
+    sessionChat: true
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - create_artifact
+    forbiddenToolsUsed:
+      - write_flow
+      - write_script
+      - deploy_workspace_item
+  judgeChecklist:
+    - saves the plan as a markdown artifact via create_artifact rather than only replying inline
+    - the artifact content has a title, a one-line summary, and three or four bullet steps for onboarding
+    - does not create a flow or script draft yet
+
+- id: global-npm1-script-search-package
+  prompt: |-
+    Find a good npm package for parsing RSS/Atom feeds and use it to create a draft Bun script
+    at `f/evals/global/fetch_feed`. The script should take a string `url` input and return the
+    feed title along with the titles of the 5 most recent items. Tell me which package you
+    picked and link its documentation. Leave it as an AI draft only; do not deploy.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json
+  runtime:
+    maxTurns: 12
+  validate:
+    draftCountExactly: 1
+    requiredDrafts:
+      - type: script
+        path: f/evals/global/fetch_feed
+        language: bun
+        valueIncludes:
+          - url
+  toolExpect:
+    requiredToolsUsed:
+      - search_npm_packages
+      - write_script
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+  judgeChecklist:
+    - searches npm for a feed-parsing package before writing the code
+    - the draft Bun script imports the chosen npm package rather than hand-rolling an XML/RSS parser
+    - the script takes a url input and returns the feed title plus the recent item titles
+    - the reply names the chosen package and links its documentation
+    - the result stays as an AI draft and is not deployed
+
+- id: global-dbschema1-postgres-resource-tables
+  prompt: |-
+    I have a postgres resource at f/data/reports_pg in this workspace.
+    What tables does that database have?
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_empty.json
+  runtime:
+    maxTurns: 12
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - get_db_schema
+    forbiddenToolsUsed:
+      - write_script
+      - write_resource
+      - test_run_script
+      - deploy_workspace_item
+    toolCallArgs:
+      - tool: get_db_schema
+        field: resourcePath
+        stringIncludesAnyOf:
+          - f/data/reports_pg
+  skipJudge: true
+  judgeChecklist:
+    - fetches the schema through get_db_schema with the resource path f/data/reports_pg
+    - when the lookup fails, tells the user instead of inventing table names
+    - does not write scripts or resources to answer a read-only question
+
+# --- API catalog (search_api_endpoints / call_api_get / call_api_endpoint) ---
+# The harness serves the catalog and the executed calls itself (mock
+# listMcpTools + benchmark fetch handlers in adapters/frontend), so these cases
+# do not require an mcp-enabled eval backend.
+
+- id: global-test30-api-catalog-workers
+  prompt: |-
+    Are any workers online right now? Give me a quick summary of the workers and when they last pinged.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_empty.json
+  runtime:
+    maxTurns: 8
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - search_api_endpoints
+      - call_api_get
+    forbiddenToolsUsed:
+      - call_api_endpoint
+      - write_script
+      - deploy_workspace_item
+    toolCallArgs:
+      - tool: call_api_get
+        field: name
+        stringIncludesAnyOf:
+          - listWorkers
+  # Read-only workspace inspection produces no draft; validate via tool use.
+  skipJudge: true
+  judgeChecklist:
+    - discovers the workers endpoint through the API catalog instead of guessing or fabricating
+    - reports worker status from the returned data
+
+- id: global-test31-draft-test-run-not-deployed
+  prompt: |-
+    Update `f/evals/global/format_greeting` so the provided name is uppercased in the greeting, then run it with name "ada" to check it works.
+    Keep it as an AI draft only; do not deploy or save it.
+  initial: ai_evals/fixtures/frontend/global/initial/format_greeting_script.json
+  runtime:
+    maxTurns: 10
+  validate:
+    draftCountExactly: 1
+    requiredDrafts:
+      - type: script
+        path: f/evals/global/format_greeting
+        language: bun
+        valueIncludes:
+          - toUpperCase
+  toolExpect:
+    requiredToolsUsed:
+      - test_run_script
+    forbiddenToolsUsed:
+      - call_api_endpoint
+      - deploy_workspace_item
+      - delete_workspace_item
+  # The judge only sees the drafts artifact and cannot observe runs, so it always
+  # docks the prompt's "run it" requirement — validate deterministically instead:
+  # draft content via valueIncludes, the test run via toolExpect (test_run_script
+  # required, call_api_endpoint forbidden).
+  skipJudge: true
+  judgeChecklist:
+    - creates an AI draft of f/evals/global/format_greeting with the name uppercased in the greeting
+    - does not deploy or save the draft
+
+- id: global-test32-delete-via-dedicated-tool
+  prompt: |-
+    Delete the deployed script `f/evals/global/send_report_digest` from the workspace — we don't use it anymore.
+  initial: ai_evals/fixtures/frontend/global/initial/report_digest_script.json
+  runtime:
+    maxTurns: 6
+  validate:
+    draftCountExactly: 0
+  toolExpect:
+    requiredToolsUsed:
+      - delete_workspace_item
+    forbiddenToolsUsed:
+      - call_api_endpoint
+      - search_api_endpoints
+      - write_script
+  # Deletion produces no draft; validate via tool use.
+  skipJudge: true
+  judgeChecklist:
+    - deletes the deployed script via delete_workspace_item rather than a raw API endpoint
+
+- id: global-draft-diff-report
+  prompt: |-
+    Update the existing workspace script at `f/evals/global/format_greeting` so the returned message ends with an exclamation mark, keeping everything else the same.
+    Then show me exactly what your draft changes compared to the deployed version.
+    Leave the result as an AI draft only; do not deploy or save it.
+  initial: ai_evals/fixtures/frontend/global/initial/format_greeting_script.json
+  runtime:
+    maxTurns: 8
+  validate:
+    draftCountExactly: 1
+    requiredDrafts:
+      - type: script
+        path: f/evals/global/format_greeting
+        language: bun
+        valueIncludes:
+          - "!"
+  toolExpect:
+    requiredToolsUsed:
+      - diff
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      - tool: diff
+        field: path
+        stringIncludesAnyOf:
+          - f/evals/global/format_greeting
+  judgeChecklist:
+    - creates an AI draft for the existing f/evals/global/format_greeting script with the exclamation-mark change
+    - the draft changes only the returned message's punctuation — summary, language, path, and the rest of the code are untouched
+    - does not deploy or save the draft to the workspace
+
+- id: global-resource-manual-credentials
+  prompt: |-
+    Set up a resource for our production Postgres database at `f/evals/global/prod_db` (host db.internal.example.com, port 5432, database `orders`, user `app`).
+    I don't want to paste the password into this chat — prepare everything so I can enter it myself.
+    Leave it as an AI draft only; do not deploy or save it.
+  initial: ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json
+  runtime:
+    maxTurns: 12
+  validate:
+    draftCountAtLeast: 1
+    requiredDrafts:
+      - type: resource
+        path: f/evals/global/prod_db
+        valueIncludes:
+          - db.internal.example.com
+          - orders
+  toolExpect:
+    requiredToolsUsed:
+      - write_resource
+      - open_page
+    forbiddenToolsUsed:
+      - deploy_workspace_item
+      - delete_workspace_item
+    toolCallArgs:
+      # The model may land the user in the resource's drawer or in the drawer of
+      # the secret variable it created for the password — both are correct.
+      - tool: open_page
+        field: page
+        stringIncludesAnyOf:
+          - resources
+          - variables
+      - tool: open_page
+        field: open
+        stringIncludesAnyOf:
+          - prod_db
+          - password
+  judgeChecklist:
+    - creates a postgres resource draft at f/evals/global/prod_db with the provided host, port, database, and user
+    - the password is left for the user to provide (empty, a placeholder, or a secret variable reference) — no invented password value presented as real
+    - does not deploy or save anything to the workspace
@@ -212,6 +212,9 @@ describe("loadCases", () => {
        },
      ],
    });
+    expect(caseEntry?.initialPath).toContain(
+      "ai_evals/fixtures/frontend/global/initial/user_admin_evals_folder.json"
+    );
    expect(caseEntry?.toolExpect).toMatchObject({
      requiredToolsUsed: ["write_script"],
      forbiddenToolsUsed: ["deploy_workspace_item", "delete_workspace_item"],
@@ -31,6 +31,8 @@ export interface EvalCaseRuntimeSpec {
  maxTurns?: number;
  backendPreview?: EvalCaseRuntimeBackendPreview;
  appContext?: EvalCaseRuntimeAppContextSpec;
+  // Global mode: run as a session chat (preview tools + session prompt) vs the standalone chat.
+  sessionChat?: boolean;
 }

 export interface FlowValidationSpec {
@@ -47,7 +49,7 @@ export interface FlowValidationSpec {
  }>;
  topLevelStepTypes?: Array<{
    id: string;
-    type: string;
+    type: string | string[];
  }>;
  moduleRules?: Array<{
    id: string;
@@ -174,6 +174,40 @@ describe("validateToolExpectations", () => {
    expect(checks.every((check) => check.passed)).toBe(true);
  });

+  it("accepts a stringIncludesAnyOf substring inside an array-valued field", () => {
+    const checks = validateToolExpectations({
+      run: {
+        success: true,
+        actual: {},
+        assistantMessageCount: 1,
+        toolCallCount: 1,
+        toolsUsed: ["open_page"],
+        toolCallDetails: [
+          {
+            name: "open_page",
+            arguments: {
+              page: "compare",
+              items: ["script:f/evals/global/compare_review_demo"],
+            },
+          },
+        ],
+        skillsInvoked: [],
+      },
+      toolExpect: {
+        requiredToolsUsed: ["open_page"],
+        toolCallArgs: [
+          {
+            tool: "open_page",
+            field: "items",
+            stringIncludesAnyOf: ["f/evals/global/compare_review_demo"],
+          },
+        ],
+      },
+    });
+
+    expect(checks.every((check) => check.passed)).toBe(true);
+  });
+
  it("accepts stringIncludesAnyOf when only one of several calls matches", () => {
    // Existential: a mutation mixed with verification SELECTs still passes.
    const checks = validateToolExpectations({
@@ -239,9 +239,15 @@ export function validateToolExpectations(input: {
      // model mixes the requested statement (e.g. an UPDATE) with verification
      // SELECTs that would otherwise fail an "all calls" check.
      const needles = rule.stringIncludesAnyOf.map((needle) => needle.toLowerCase());
-      const hasMatch = values.some(
-        (value) =>
-          typeof value === "string" && needles.some((needle) => value.toLowerCase().includes(needle))
+      // Array-valued fields (e.g. open_page.items) match on any element.
+      const haystacks = (value: unknown): string[] =>
+        typeof value === "string"
+          ? [value]
+          : Array.isArray(value)
+            ? value.filter((v): v is string => typeof v === "string")
+            : [];
+      const hasMatch = values.some((value) =>
+        haystacks(value).some((hay) => needles.some((needle) => hay.toLowerCase().includes(needle)))
      );
      checks.push(
        check(
@@ -1378,11 +1384,14 @@ function validateFlowRequirements(
      continue;
    }

+    const allowedTypes = Array.isArray(requiredStep.type)
+      ? requiredStep.type
+      : [requiredStep.type];
    checks.push(
      check(
        `${requiredStep.id} type matches required`,
-        getModuleType(module) === requiredStep.type,
-        `expected ${requiredStep.type}, got ${getModuleType(module) ?? "(missing)"}`
+        allowedTypes.includes(getModuleType(module) ?? ""),
+        `expected ${allowedTypes.join(" or ")}, got ${getModuleType(module) ?? "(missing)"}`
      )
    );
  }
@@ -0,0 +1,8 @@
+{
+  "user": {
+    "username": "admin",
+    "is_admin": true,
+    "folders": ["evals"],
+    "folders_read": ["evals"]
+  }
+}
@@ -41,6 +41,7 @@ export function createGlobalModeRunner(
          workspaceFixtures: initial?.workspace,
          liveEditorDrafts: initial?.liveEditorDrafts,
          user: initial?.user,
+          sessionChat: context.evalCase?.runtime?.sessionChat,
          maxIterations: context.evalCase?.runtime?.maxTurns,
          provider: modelConfig.provider,
          model: modelConfig.model,
@@ -0,0 +1,15 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "DELETE FROM datatable_migrations WHERE workspace_id = $1 AND datatable = ANY($2::text[])",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Text",
+        "TextArray"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "00e9fc3fed9379264881c58df9404ab3569a0611b33c261602d62e9c847c583e"
+}
@@ -0,0 +1,33 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "UPDATE materialized_asset_schema\n                    SET snapshot_id = $5, job_id = $6, captured_at = now()\n                  WHERE workspace_id = $1 AND asset_kind = $2 AND asset_path = $3\n                    AND version = $4",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Text",
+        {
+          "Custom": {
+            "name": "asset_kind",
+            "kind": {
+              "Enum": [
+                "s3object",
+                "resource",
+                "variable",
+                "ducklake",
+                "datatable",
+                "volume"
+              ]
+            }
+          }
+        },
+        "Text",
+        "Int8",
+        "Int8",
+        "Uuid"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "01732ca02b1888145c48c4e51e5b5829657224a743af9c0b2d5a140ad70e13dd"
+}
@@ -0,0 +1,22 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT args->'__git_sync_auto_pull' FROM v2_job WHERE id = $1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "?column?",
+        "type_info": "Jsonb"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Uuid"
+      ]
+    },
+    "nullable": [
+      null
+    ]
+  },
+  "hash": "020cc73e39782e7be50a146373096bd4378988af6447feeafd243262c246a9f5"
+}
@@ -0,0 +1,38 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "\n            INSERT INTO amqp_trigger (\n                amqp_resource_path,\n                queue_name,\n                exchange,\n                options,\n                workspace_id,\n                path,\n                script_path,\n                is_flow,\n                permissioned_as,\n                mode,\n                edited_by,\n                error_handler_path,\n                error_handler_args,\n                retry\n            )\n            VALUES (\n                $1, $2, $3, $4, $5, $6, $7, $8, $9, $10, $11, $12, $13, $14\n            )",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Varchar",
+        "Jsonb",
+        "Jsonb",
+        "Varchar",
+        "Varchar",
+        "Varchar",
+        "Bool",
+        "Varchar",
+        {
+          "Custom": {
+            "name": "trigger_mode",
+            "kind": {
+              "Enum": [
+                "enabled",
+                "disabled",
+                "suspended"
+              ]
+            }
+          }
+        },
+        "Varchar",
+        "Varchar",
+        "Jsonb",
+        "Jsonb"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "02e39bb9957d3ccfe3d573994cd281a22467010d2c4a0611c8ae02742336f433"
+}
@@ -1,38 +0,0 @@
-{
-  "db_name": "PostgreSQL",
-  "query": "\n        DELETE FROM asset\n        WHERE (workspace_id, path, kind) IN (\n            SELECT workspace_id, path, kind FROM (\n                SELECT a.workspace_id, a.path, a.kind, a.usage_kind, ROW_NUMBER() OVER (\n                    PARTITION BY a.workspace_id, a.path, a.kind\n                    ORDER BY a.created_at DESC\n                ) as rn,\n                limits.max_n\n                FROM asset a\n                INNER JOIN (\n                    SELECT * FROM UNNEST(\n                        $1::varchar[], \n                        $2::varchar[], \n                        $3::asset_kind[],\n                        $4::int[]\n                    ) AS t(workspace_id, path, kind, max_n)\n                ) limits\n                ON a.workspace_id = limits.workspace_id \n                AND a.path = limits.path \n                AND a.kind = limits.kind\n                WHERE a.usage_kind = 'job'\n            ) ranked\n            WHERE rn > max_n\n        )",
-  "describe": {
-    "columns": [],
-    "parameters": {
-      "Left": [
-        "VarcharArray",
-        "VarcharArray",
-        {
-          "Custom": {
-            "name": "asset_kind[]",
-            "kind": {
-              "Array": {
-                "Custom": {
-                  "name": "asset_kind",
-                  "kind": {
-                    "Enum": [
-                      "s3object",
-                      "resource",
-                      "variable",
-                      "ducklake",
-                      "datatable",
-                      "volume"
-                    ]
-                  }
-                }
-              }
-            }
-          }
-        },
-        "Int4Array"
-      ]
-    },
-    "nullable": []
-  },
-  "hash": "02e526146f3584cd599dec708e1be48db3b0cd1c74adbfa2e4039377daa016f0"
-}
@@ -0,0 +1,12 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "DELETE FROM feature_usage WHERE day < CURRENT_DATE - 60",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": []
+    },
+    "nullable": []
+  },
+  "hash": "0390d9e4fae597aabdeef940aa5b5014e1c889ba9181ee3331f128c56c9d22de"
+}
@@ -0,0 +1,28 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT workspace_id as \"workspace_id!\", MIN(completed_at) as oldest\n             FROM v2_job_completed\n             WHERE workspace_id = ANY($1::text[])\n             GROUP BY workspace_id",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "workspace_id!",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "oldest",
+        "type_info": "Timestamptz"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "TextArray"
+      ]
+    },
+    "nullable": [
+      false,
+      null
+    ]
+  },
+  "hash": "056231b872a080a257c034489127e253678631e091b9e4223842648a4475dbbf"
+}
@@ -0,0 +1,41 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "\n        SELECT kind, path, script_path, is_flow FROM (\n            SELECT 'schedule' AS kind, path, script_path, is_flow FROM schedule\n                WHERE workspace_id = $1\n                  AND script_path IS NOT NULL\n            UNION ALL\n            SELECT 'email', path, script_path, is_flow FROM email_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'kafka', path, script_path, is_flow FROM kafka_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'mqtt', path, script_path, is_flow FROM mqtt_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'amqp', path, script_path, is_flow FROM amqp_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'nats', path, script_path, is_flow FROM nats_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'postgres', path, script_path, is_flow FROM postgres_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'sqs', path, script_path, is_flow FROM sqs_trigger\n                WHERE workspace_id = $1\n            UNION ALL\n            SELECT 'gcp', path, script_path, is_flow FROM gcp_trigger\n                WHERE workspace_id = $1\n        ) t\n        WHERE ($2::text IS NULL OR script_path LIKE $2)\n        ",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "kind",
+        "type_info": "Text"
+      },
+      {
+        "ordinal": 1,
+        "name": "path",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 2,
+        "name": "script_path",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 3,
+        "name": "is_flow",
+        "type_info": "Bool"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": [
+      null,
+      null,
+      null,
+      null
+    ]
+  },
+  "hash": "05843e117361d6b7cd9da652596d3be7a970d602d3809396d9253d7e5dc88152"
+}
@@ -0,0 +1,15 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "UPDATE workspace_settings SET deploy_to = $1 WHERE deploy_to = $2",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Text"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "0621faf69b1ef866a95f6310c9651875df409a2d2d72ada629bd71e8abdbbf8b"
+}
@@ -0,0 +1,20 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT SUM(pg_database_size(datname))::BIGINT AS \"v!\" FROM pg_database",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "v!",
+        "type_info": "Int8"
+      }
+    ],
+    "parameters": {
+      "Left": []
+    },
+    "nullable": [
+      null
+    ]
+  },
+  "hash": "074dd26f6427f4ff97e92c35163ad042144e656adcca56a4936a2eb196d3f48c"
+}
@@ -0,0 +1,28 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT parent_workspace_id, is_dev_workspace FROM workspace WHERE id = $1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "parent_workspace_id",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "is_dev_workspace",
+        "type_info": "Bool"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      true,
+      false
+    ]
+  },
+  "hash": "079e1dab42a783a1e5e4cba5faa854b4b727b1fab89e42f2efaf97c63dad6384"
+}
@@ -35,7 +35,9 @@
                "ci_test",
                "github",
                "azure",
-                "asset"
+                "asset",
+                "freshness",
+                "amqp"
              ]
            }
          }
@@ -0,0 +1,28 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "\n        SELECT\n            (elem->>'installation_id')::bigint as installation_id,\n            elem->>'github_base_url' as github_base_url\n        FROM workspace_settings,\n        LATERAL jsonb_array_elements(git_app_installations) AS elem\n        WHERE workspace_id = $1\n        ",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "installation_id",
+        "type_info": "Int8"
+      },
+      {
+        "ordinal": 1,
+        "name": "github_base_url",
+        "type_info": "Text"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      null,
+      null
+    ]
+  },
+  "hash": "0bb3de62f920a4f6c5da6f0a7bfc780a5ac35fa7639c4ee3a449b693add24241"
+}
@@ -0,0 +1,14 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO v2_job (id, workspace_id, tag) VALUES ($1, 'test-workspace', 'other')",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Uuid"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "0bd853be57b43a7820cc1d73d92b8babd50b1092d4ad70e022431bdcf32e378f"
+}
@@ -0,0 +1,34 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT\n                (SELECT MIN(completed_at) FROM v2_job_completed) as true_oldest,\n                (SELECT MIN(completed_at) FROM v2_job_completed\n                 WHERE workspace_id <> ALL($1::text[])) as global_oldest,\n                (SELECT COUNT(*) FROM v2_job_completed) as total",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "true_oldest",
+        "type_info": "Timestamptz"
+      },
+      {
+        "ordinal": 1,
+        "name": "global_oldest",
+        "type_info": "Timestamptz"
+      },
+      {
+        "ordinal": 2,
+        "name": "total",
+        "type_info": "Int8"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "TextArray"
+      ]
+    },
+    "nullable": [
+      null,
+      null,
+      null
+    ]
+  },
+  "hash": "0d373b2600273763c002b1e75f743e1b981f7884df99f033838c4336bea2b051"
+}
@@ -0,0 +1,23 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT args->>$2 FROM v2_job WHERE id = ANY($1)",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "?column?",
+        "type_info": "Text"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "UuidArray",
+        "Text"
+      ]
+    },
+    "nullable": [
+      null
+    ]
+  },
+  "hash": "0d6c3399bc637e2c599534c2cafacf623cd6e5fde2d075a4f1f0d4852efcedca"
+}
@@ -0,0 +1,24 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "WITH RECURSIVE descendants AS (\n                   SELECT id, 0 AS depth FROM workspace\n                   WHERE parent_workspace_id = $1 AND NOT deleted\n                   UNION ALL\n                   SELECT w.id, d.depth + 1 FROM workspace w\n                   JOIN descendants d ON w.parent_workspace_id = d.id\n                   WHERE NOT w.deleted AND d.depth < 10\n               )\n               SELECT id as \"id!\" FROM descendants WHERE (id = $2 OR id = $3)\n               ORDER BY (id = $2) DESC LIMIT 1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "id!",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": [
+      null
+    ]
+  },
+  "hash": "0e016aaa6842767ac636e5b3e1febfa9cee08b93e0562beb629d84c4c689f4e7"
+}
@@ -0,0 +1,15 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "DELETE FROM pipeline_freshness_state WHERE workspace_id = $1 AND script_path = $2",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "0e22295e40bfc173d85f0a962d491f3afd87eeea5e4e8d935b1b86eb02f061e7"
+}
@@ -0,0 +1,77 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT DISTINCT ON (mp.asset_kind, mp.asset_path)\n                  mp.asset_kind AS \"asset_kind: AssetKind\", mp.asset_path,\n                  mp.snapshot_id AS \"snapshot_id!\", mp.partition\n             FROM materialized_partition mp\n             JOIN unnest($2::ASSET_KIND[], $3::text[]) AS u(kind, path)\n               ON mp.asset_kind = u.kind AND mp.asset_path = u.path\n            WHERE mp.workspace_id = $1\n              AND mp.status = 'materialized' AND mp.snapshot_id IS NOT NULL\n            ORDER BY mp.asset_kind, mp.asset_path, mp.snapshot_id DESC",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "asset_kind: AssetKind",
+        "type_info": {
+          "Custom": {
+            "name": "asset_kind",
+            "kind": {
+              "Enum": [
+                "s3object",
+                "resource",
+                "variable",
+                "ducklake",
+                "datatable",
+                "volume"
+              ]
+            }
+          }
+        }
+      },
+      {
+        "ordinal": 1,
+        "name": "asset_path",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 2,
+        "name": "snapshot_id!",
+        "type_info": "Int8"
+      },
+      {
+        "ordinal": 3,
+        "name": "partition",
+        "type_info": "Text"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        {
+          "Custom": {
+            "name": "asset_kind[]",
+            "kind": {
+              "Array": {
+                "Custom": {
+                  "name": "asset_kind",
+                  "kind": {
+                    "Enum": [
+                      "s3object",
+                      "resource",
+                      "variable",
+                      "ducklake",
+                      "datatable",
+                      "volume"
+                    ]
+                  }
+                }
+              }
+            }
+          }
+        },
+        "TextArray"
+      ]
+    },
+    "nullable": [
+      false,
+      false,
+      true,
+      false
+    ]
+  },
+  "hash": "0ee4a4dc8c6f09e7027bd99f98d8d1d346f9d67056ef272314dd9b80bee9f285"
+}
@@ -0,0 +1,31 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT\n            schedule.path, t.jobs FROM schedule,\n            LATERAL(SELECT ARRAY(\n                SELECT json_build_object('id', id, 'success', status = 'success', 'duration_ms', duration_ms)\n                FROM v2_job_completed c JOIN v2_job j USING (id)\n                WHERE trigger_kind = 'schedule'\n                    AND trigger = schedule.path\n                    AND c.workspace_id = $1\n                    AND j.workspace_id = $1\n                    AND parent_job IS NULL AND runnable_path = schedule.script_path\n                    AND status <> 'skipped'\n                ORDER BY completed_at DESC\n                LIMIT 20\n            ) AS jobs) t\n        WHERE workspace_id = $1 AND NOT starts_with(schedule.path, $4)\n        ORDER BY edited_at DESC\n        LIMIT $2 OFFSET $3",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "path",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "jobs",
+        "type_info": "JsonArray"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Int8",
+        "Int8",
+        "Text"
+      ]
+    },
+    "nullable": [
+      false,
+      null
+    ]
+  },
+  "hash": "0f141ca6a58901dee1ddcf25694d76a4e5702c0324a27ef3a5740ef5d98d8946"
+}
@@ -27,7 +27,8 @@
                "nextcloud",
                "google",
                "github",
-                "azure"
+                "azure",
+                "amqp"
              ]
            }
          }
@@ -0,0 +1,12 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO script (workspace_id, path, hash, content, summary, description, language, created_by, created_at, archived, schema_validation, ws_error_handler_muted, deleted, extra_perms)\n         VALUES ('test-workspace', 'f/restricted/item', 314159, 'def main(): return 1', '', '', 'python3', 'test-user', NOW(), false, false, false, false, '{}'::jsonb)",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": []
+    },
+    "nullable": []
+  },
+  "hash": "0ff0b7abad7717d7de398f629853b22cb831e91026dcd59d362a6d4382dc240b"
+}
@@ -0,0 +1,58 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "\n        SELECT language AS \"language!: _\", COUNT(*)::BIGINT AS \"count!\"\n        FROM script\n        WHERE archived = false AND deleted = false AND kind = 'script'\n          AND (auto_kind IS NULL OR auto_kind <> 'wac')\n        GROUP BY language\n        ",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "language!: _",
+        "type_info": {
+          "Custom": {
+            "name": "script_lang",
+            "kind": {
+              "Enum": [
+                "python3",
+                "deno",
+                "go",
+                "bash",
+                "postgresql",
+                "nativets",
+                "bun",
+                "mysql",
+                "bigquery",
+                "snowflake",
+                "graphql",
+                "powershell",
+                "mssql",
+                "php",
+                "bunnative",
+                "rust",
+                "ansible",
+                "csharp",
+                "oracledb",
+                "nu",
+                "java",
+                "duckdb",
+                "ruby",
+                "rlang"
+              ]
+            }
+          }
+        }
+      },
+      {
+        "ordinal": 1,
+        "name": "count!",
+        "type_info": "Int8"
+      }
+    ],
+    "parameters": {
+      "Left": []
+    },
+    "nullable": [
+      false,
+      null
+    ]
+  },
+  "hash": "11813108dbf6b104eba968c3609c74ac5a589542d3e76b1e756a82fb19d49ee8"
+}
@@ -0,0 +1,46 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT name, params, body, is_table_macro, provider_path FROM macro_definition WHERE workspace_id = $1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "name",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "params",
+        "type_info": "Text"
+      },
+      {
+        "ordinal": 2,
+        "name": "body",
+        "type_info": "Text"
+      },
+      {
+        "ordinal": 3,
+        "name": "is_table_macro",
+        "type_info": "Bool"
+      },
+      {
+        "ordinal": 4,
+        "name": "provider_path",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      false,
+      false,
+      false,
+      false,
+      false
+    ]
+  },
+  "hash": "12dec9fa086fd231cdba2ecaa4585b62e55f43d448332ff88d8f036443b1edd7"
+}
@@ -0,0 +1,16 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO schedule (\n            workspace_id, path, edited_by, edited_at, schedule, enabled, script_path,\n            args, extra_perms, is_flow, email, error, timezone, on_failure,\n            on_recovery, on_failure_times, on_failure_exact, on_failure_extra_args,\n            on_recovery_times, on_recovery_extra_args, ws_error_handler_muted, retry,\n            summary, no_flow_overlap, tag, paused_until, on_success, on_success_extra_args,\n            cron_version, description, dynamic_skip, permissioned_as, labels\n        )\n        SELECT\n            $1, path, edited_by, edited_at, schedule, FALSE, script_path,\n            args, extra_perms, is_flow, email, error, timezone, on_failure,\n            on_recovery, on_failure_times, on_failure_exact, on_failure_extra_args,\n            on_recovery_times, on_recovery_extra_args, ws_error_handler_muted, retry,\n            summary, no_flow_overlap, tag, paused_until, on_success, on_success_extra_args,\n            cron_version, description, dynamic_skip, permissioned_as, labels\n        FROM schedule WHERE workspace_id = $2 AND path NOT LIKE $3",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "142ae6939440654da21ebb06acf838563f28e832e37e5c165329441d8471acdb"
+}
@@ -1,26 +0,0 @@
-{
-  "db_name": "PostgreSQL",
-  "query": "\n        WITH completed AS (\n            INSERT INTO v2_job_completed\n                (workspace_id, id, started_at, duration_ms, result,\n                 flow_status, workflow_as_code_status, status, worker)\n            SELECT\n                q.workspace_id, q.id, q.started_at,\n                (EXTRACT('epoch' FROM now()) - EXTRACT('epoch' FROM COALESCE(q.started_at, now()))) * 1000,\n                CASE WHEN q.running\n                    THEN $3::text::jsonb\n                    ELSE $4::text::jsonb\n                END,\n                s.flow_status,\n                s.workflow_as_code_status,\n                'skipped'::job_status,\n                q.worker\n            FROM v2_job_queue q\n            LEFT JOIN v2_job_status s ON s.id = q.id\n            WHERE q.id = $1\n            ON CONFLICT (id) DO UPDATE SET status = EXCLUDED.status, result = EXCLUDED.result\n            RETURNING 1 AS x\n        ), _deleted AS (\n            DELETE FROM v2_job_queue WHERE id = $1\n        ), _logged AS (\n            INSERT INTO job_logs (logs, job_id, workspace_id)\n            VALUES ($5, $1, $2)\n            ON CONFLICT (job_id) DO UPDATE SET logs = concat(job_logs.logs, EXCLUDED.logs)\n        )\n        SELECT x FROM completed\n        ",
-  "describe": {
-    "columns": [
-      {
-        "ordinal": 0,
-        "name": "x",
-        "type_info": "Int4"
-      }
-    ],
-    "parameters": {
-      "Left": [
-        "Uuid",
-        "Varchar",
-        "Text",
-        "Text",
-        "Text"
-      ]
-    },
-    "nullable": [
-      null
-    ]
-  },
-  "hash": "1437b432d2c23e30eb05443e83069cdb049f65ec299b0778ce14677728cf6346"
-}
@@ -0,0 +1,23 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT EXISTS(SELECT 1 FROM datatable_migrations WHERE workspace_id = $1 AND datatable = $2)",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "exists",
+        "type_info": "Bool"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": [
+      null
+    ]
+  },
+  "hash": "15c0584e9eb078442f6928adb894324c47e010e3554c14e755a7e045453745bb"
+}
@@ -0,0 +1,28 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT id, name FROM workspace WHERE parent_workspace_id = $1 AND is_dev_workspace AND deleted = false",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "id",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "name",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      false,
+      false
+    ]
+  },
+  "hash": "165abf847773c7fe718fa6c832b259bf2dc531bc468d2ab27f76c9c653be0b9a"
+}
@@ -1,22 +0,0 @@
-{
-  "db_name": "PostgreSQL",
-  "query": "\n                SELECT ws.datatable->'datatables' AS datatable_name\n                FROM workspace_settings ws\n                WHERE ws.workspace_id = $1\n            ",
-  "describe": {
-    "columns": [
-      {
-        "ordinal": 0,
-        "name": "datatable_name",
-        "type_info": "Jsonb"
-      }
-    ],
-    "parameters": {
-      "Left": [
-        "Text"
-      ]
-    },
-    "nullable": [
-      null
-    ]
-  },
-  "hash": "16a67b92dbd32024838983184e6974f2ce577b78decd6b821096c9f2f252ae8b"
-}
@@ -0,0 +1,28 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT (parent_workspace_id IS NOT NULL) AS \"has_parent!\", is_dev_workspace\n           FROM workspace WHERE id = $1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "has_parent!",
+        "type_info": "Bool"
+      },
+      {
+        "ordinal": 1,
+        "name": "is_dev_workspace",
+        "type_info": "Bool"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      null,
+      false
+    ]
+  },
+  "hash": "16c7838ecfcea5fd231f2a4766f691a9a11ca7bb9797b81444419d6a72883531"
+}
@@ -27,7 +27,8 @@
                "nextcloud",
                "google",
                "github",
-                "azure"
+                "azure",
+                "amqp"
              ]
            }
          }
@@ -0,0 +1,26 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT COALESCE(username, split_part(email, '@', 1)) AS \"username!\", email FROM password WHERE super_admin = true AND disabled = false ORDER BY email LIMIT 1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "username!",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "email",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": []
+    },
+    "nullable": [
+      null,
+      false
+    ]
+  },
+  "hash": "17cdf02b4912078459526205849246fd2bdf9e3fce89852120aa4ad4ad6abfc3"
+}
@@ -0,0 +1,18 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO workspace_multipart_inflight\n           (workspace_id, upload_id, storage, inflight_bytes, target_existing_size)\n         VALUES ($1, $2, $3, $4, $5)\n         ON CONFLICT (workspace_id, upload_id)\n         DO UPDATE SET inflight_bytes = $4",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Varchar",
+        "Varchar",
+        "Int8",
+        "Int8"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "1b05728b33decc39766ccacca50464b5ff9f34e43fbcc38154939461aadfca9f"
+}
@@ -0,0 +1,23 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT id FROM workspace WHERE parent_workspace_id = $1 AND NOT deleted AND is_dev_workspace AND COALESCE(dev_workspace_label, 'dev') = $2",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "id",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": [
+      false
+    ]
+  },
+  "hash": "1b32339718406531acd2d051386d2817e9ba0b3a66f7eb4fc1177764563a0665"
+}
@@ -0,0 +1,46 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT w.parent_workspace_id, w.is_dev_workspace, w.dev_workspace_label,\n                  p.is_dev_workspace as \"parent_is_dev_workspace?\", p.dev_workspace_label as \"parent_dev_label?\"\n           FROM workspace w LEFT JOIN workspace p ON p.id = w.parent_workspace_id\n           WHERE w.id = $1",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "parent_workspace_id",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "is_dev_workspace",
+        "type_info": "Bool"
+      },
+      {
+        "ordinal": 2,
+        "name": "dev_workspace_label",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 3,
+        "name": "parent_is_dev_workspace?",
+        "type_info": "Bool"
+      },
+      {
+        "ordinal": 4,
+        "name": "parent_dev_label?",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": [
+      true,
+      false,
+      true,
+      false,
+      true
+    ]
+  },
+  "hash": "1c018b483795d9d33baa7be2ee41f48da6fe47bc29b6987cf60edd56bb1663b6"
+}
@@ -1,27 +0,0 @@
-{
-  "db_name": "PostgreSQL",
-  "query": "DELETE FROM asset WHERE workspace_id = $1 AND usage_path = $2 AND usage_kind = $3",
-  "describe": {
-    "columns": [],
-    "parameters": {
-      "Left": [
-        "Text",
-        "Text",
-        {
-          "Custom": {
-            "name": "asset_usage_kind",
-            "kind": {
-              "Enum": [
-                "script",
-                "flow",
-                "job"
-              ]
-            }
-          }
-        }
-      ]
-    },
-    "nullable": []
-  },
-  "hash": "1c5caaaa86e3488549cad179e992172315be5d53dcc266d713da01fd27f310b6"
-}
@@ -0,0 +1,17 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "WITH t1 AS (UPDATE websocket_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t2 AS (UPDATE kafka_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t3 AS (UPDATE postgres_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t4 AS (UPDATE mqtt_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t5 AS (UPDATE nats_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t6 AS (UPDATE sqs_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4), t7 AS (UPDATE amqp_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4) UPDATE gcp_trigger SET script_path = $1, server_id = NULL WHERE script_path = $2 AND workspace_id = $3 AND is_flow = $4",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Text",
+        "Text",
+        "Bool"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "1cd16f93bdc90b24ceb1e86495069c3592218d34e58d617253e8b563bc661a7a"
+}
@@ -0,0 +1,18 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "\n            INSERT INTO workspace_protection_rule (workspace_id, name, rules, bypass_groups, bypass_users)\n            VALUES ($1, $2, $3, $4, $5)\n            ON CONFLICT (workspace_id, name)\n            DO UPDATE SET rules = EXCLUDED.rules,\n                          bypass_groups = EXCLUDED.bypass_groups,\n                          bypass_users = EXCLUDED.bypass_users\n        ",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Varchar",
+        "Varchar",
+        "Int4",
+        "TextArray",
+        "TextArray"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "1cd7c77e7a6a5c13c4ca521098bb07c1d805d21899fe0ebac22132b248ffd242"
+}
@@ -0,0 +1,17 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO v2_job_queue (id, workspace_id, scheduled_for, tag, running)\n             VALUES ($1, $2, now() + ($3::bigint::text || ' s')::interval, 'other', $4)",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Uuid",
+        "Varchar",
+        "Int8",
+        "Bool"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "1ce0728a6b0942fecf8ce5f4a06857a2601b10b057877c3b2795b8ea9c928c1b"
+}
@@ -1,12 +1,17 @@
 {
  "db_name": "PostgreSQL",
-  "query": "SELECT restart_unless_cancelled FROM script WHERE hash = $1 AND workspace_id = $2",
+  "query": "SELECT restart_unless_cancelled, timeout FROM script WHERE hash = $1 AND workspace_id = $2",
  "describe": {
    "columns": [
      {
        "ordinal": 0,
        "name": "restart_unless_cancelled",
        "type_info": "Bool"
+      },
+      {
+        "ordinal": 1,
+        "name": "timeout",
+        "type_info": "Int4"
      }
    ],
    "parameters": {
@@ -16,8 +21,9 @@
      ]
    },
    "nullable": [
+      true,
      true
    ]
  },
-  "hash": "1d27895aa42ccbb542479b19baefd62790205b529ab0d8af36f18c470e8bb838"
+  "hash": "1debd472c9ffd2fc78877484f93db51f9aabed54f9894eda8ad610053ad76ce6"
 }
@@ -0,0 +1,16 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "INSERT INTO v2_job_queue (id, workspace_id, scheduled_for, tag, running)\n             VALUES ($1, 'test-workspace', now() + ($2::bigint::text || ' s')::interval, 'other', $3)",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Uuid",
+        "Int8",
+        "Bool"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "20ae664da5ebd70a6d37609f6b9a3e10d8d524fbaaa8eddf74f12a906b3c8306"
+}
@@ -0,0 +1,32 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "DELETE FROM v2_job_completed\n                 WHERE id IN (\n                     SELECT jc.id FROM v2_job_completed jc\n                     LEFT JOIN v2_job j ON j.id = jc.id\n                     WHERE jc.workspace_id = $5\n                       AND jc.completed_at <= now() - ($1::bigint::text || ' s')::interval\n                       AND jc.completed_at >= COALESCE($4::timestamptz, '-infinity'::timestamptz)\n                       AND COALESCE(j.root_job, j.flow_innermost_root_job, jc.id) NOT IN (\n                           SELECT u FROM unnest($3::uuid[]) AS u WHERE u IS NOT NULL\n                       )\n                     ORDER BY jc.completed_at ASC\n                     LIMIT $2\n                     FOR UPDATE OF jc SKIP LOCKED\n                 )\n                 RETURNING id, completed_at",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "id",
+        "type_info": "Uuid"
+      },
+      {
+        "ordinal": 1,
+        "name": "completed_at",
+        "type_info": "Timestamptz"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Int8",
+        "Int8",
+        "UuidArray",
+        "Timestamptz",
+        "Text"
+      ]
+    },
+    "nullable": [
+      false,
+      false
+    ]
+  },
+  "hash": "20d62e1937c855cc6887597b3860e991cb539320f4c8321af94416f702108e2e"
+}
@@ -0,0 +1,29 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "SELECT consumer_path AS \"consumer_path!\", macro_name AS \"macro_name!\"\n           FROM macro_usage\n           WHERE workspace_id = $1\n             AND ($2::text IS NULL OR consumer_path LIKE $2)",
+  "describe": {
+    "columns": [
+      {
+        "ordinal": 0,
+        "name": "consumer_path!",
+        "type_info": "Varchar"
+      },
+      {
+        "ordinal": 1,
+        "name": "macro_name!",
+        "type_info": "Varchar"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        "Text"
+      ]
+    },
+    "nullable": [
+      false,
+      false
+    ]
+  },
+  "hash": "21242859ee6b72466c6bfb9d826a7159de4488adccb9349cddfbdd572b4005e9"
+}
@@ -38,7 +38,10 @@
                "google",
                "ci_test",
                "github",
-                "azure"
+                "azure",
+                "asset",
+                "freshness",
+                "amqp"
              ]
            }
          }
@@ -75,7 +78,10 @@
                "google",
                "ci_test",
                "github",
-                "azure"
+                "azure",
+                "asset",
+                "freshness",
+                "amqp"
              ]
            }
          }
@@ -27,7 +27,8 @@
                "nextcloud",
                "google",
                "github",
-                "azure"
+                "azure",
+                "amqp"
              ]
            }
          }
@@ -1,12 +1,23 @@
 {
  "db_name": "PostgreSQL",
-  "query": "\n        SELECT\n            kind AS \"kind!: AssetKind\",\n            path AS \"path!\"\n        FROM asset\n        WHERE workspace_id = $1\n          AND usage_kind = 'script'\n          AND usage_path = $2\n          AND usage_access_type IN ('w', 'rw')\n        ",
+  "query": "SELECT version, columns AS \"columns: Json<Vec<SchemaColumn>>\"\n             FROM materialized_asset_schema\n            WHERE workspace_id = $1 AND asset_kind = $2 AND asset_path = $3\n            ORDER BY version DESC\n            LIMIT 1",
  "describe": {
    "columns": [
      {
        "ordinal": 0,
-        "name": "kind!: AssetKind",
-        "type_info": {
+        "name": "version",
+        "type_info": "Int8"
+      },
+      {
+        "ordinal": 1,
+        "name": "columns: Json<Vec<SchemaColumn>>",
+        "type_info": "Jsonb"
+      }
+    ],
+    "parameters": {
+      "Left": [
+        "Text",
+        {
          "Custom": {
            "name": "asset_kind",
            "kind": {
@@ -20,17 +31,7 @@
              ]
            }
          }
-        }
-      },
-      {
-        "ordinal": 1,
-        "name": "path!",
-        "type_info": "Varchar"
-      }
-    ],
-    "parameters": {
-      "Left": [
-        "Text",
+        },
        "Text"
      ]
    },
@@ -39,5 +40,5 @@
      false
    ]
  },
-  "hash": "de06f44bad94710f14e9be4c0a6e6080e3c4faae5052500b93cc24b6fe556f2b"
+  "hash": "231475dc825518aa88f562698f8061d2562c3ac4af8d088f9eea9f0bc11d5fe7"
 }
@@ -43,7 +43,8 @@
                "nextcloud",
                "google",
                "github",
-                "azure"
+                "azure",
+                "amqp"
              ]
            }
          }
@@ -0,0 +1,14 @@
+{
+  "db_name": "PostgreSQL",
+  "query": "DELETE FROM macro_definition WHERE workspace_id = $1",
+  "describe": {
+    "columns": [],
+    "parameters": {
+      "Left": [
+        "Text"
+      ]
+    },
+    "nullable": []
+  },
+  "hash": "246e302feda95892d5eb1a43a29d4d4739fd663cf34f00d947f24b47c481f035"
+}
--- a/Show More
+++ b/Show More
				`@@ -0,0 +1 @@`
				`../../../.agents/skills/local-review-codex/SKILL.md`