Files
orca/src/shared/agent-title-status.ts
T
Brennan Benson 761d63a4e5 feat(agent-launch): keep long prompts off the launch line and paste them after readiness (step 1 of 7) (#24257)
* feat(agent-launch): host-side prompt delivery for agent.launch

The host's agent.launch typed any launch prompt into the shell as part of the
launch command. A long or multi-line prompt then ran line by line in the
shell, and an agent that never showed readiness or crashed at startup had
nothing guarding where its text went.

agent.launch now carries a prompt on the typed line only when the line stays
one line, control-free and at most 512 bytes; otherwise the agent starts
clean and the host pastes the prompt once the agent's own ready signal fires
(bracketed paste plus its composer marker or a quiet render, read only after
the shell's last hand-off, never while the pane's own shell is proven in
front), with main's draft-paste bytes and an Enter 50 ms later. Orchestration
worker starts wait on tui-idle as before. A replay-safe launch admits and
claims its ledger row in one write, Qwen Code gets a second Enter, the
desktop and phone share one launch-refusal classifier, and hosts advertise
agent.launch.prompt-carry.v1.

Split out of #23748, which moves the desktop source-control buttons onto
this path.

* fix(agent-launch): keep a short-lined multi-line prompt on a local zsh launch line, as main did

#24257 moved every multi-line or over-512-byte prompt off the typed launch line and pasted it after readiness. The phone's AI buttons and review notes, whose multi-line prompts main typed whole into zsh, then reached Claude 0.5-3 s later and their RPC reply waited for the paste.

The host now names the shell a local macOS or Linux line is typed into, the way the spawn picks it, and a multi-line prompt rides a zsh line when every line is at most 512 bytes and the whole line at most 8 KB. A real-zsh test types such a line through Orca's own ready barrier and startup write, including when a slow user config makes the write land early. Elsewhere the measured unsafe cases keep the paste: bash 3.2 runs multi-line lines piecemeal, fish drops an early multi-line write, and any shell loses a line over 1 KB written early.

* test(agent-launch): keep the real-zsh launch-line test out of the Windows lane's gate scan

The Windows lane registration check read `const ZSH_PATH = process.platform === 'win32'` (the head of a multi-line ternary) as a Windows-true flag, so `describe.skipIf(!ZSH_PATH)` looked like a Windows-only suite. The file is POSIX-only; the zsh lookup is now a function.

* refactor(protocol): move the agent.launch capabilities into their own module

Main's protocol-version.ts sits at the 300-line cap, so the prompt-carry capability pushed it over. The four agent.launch capabilities and their doc move to agent-launch-runtime-capability.ts, re-exported by name and spread into RUNTIME_CAPABILITIES at the same position; the advertised lists and every export are unchanged.

* refactor(protocol): import the agent.launch capabilities from their own module

`export *` from protocol-version left the four names undefined under the mobile recording loader, which resolves a relative import through a Proxy with no own keys, so 37 phone recordings lost agent.launchReplay. Importers now name agent-launch-runtime-capability directly; protocol-version only spreads its list.

* refactor(agent-launch): drop the unshipped viewMode field and trusted local caller id

Both were inert in step 1 and existed only for step 2. agent.launch will become a
public plugin API, so every wire field is permanent once shipped; a top-level
viewMode reads as "choose terminal vs chat", which the host decides. Step 2
introduces placement and view intent under a placement object instead.

* fix(agent-launch): read Codex's provisional startup from the rule files' hold anchor

Main (#24375) moved Codex's provisional-header check into codex.json's
provisional_startup hold anchor and deleted codex-terminal-readiness.ts, so the
launch readiness hold now asks showsHoldAnchor, as main's own settled check does.

* fix(agent-launch): hold rule-file name titles to quiet for a launch, and census the zsh fixture

Main (#24375) answers a name-only title from each agent's rule file ahead of the
sustained-title lane, so gemini.json's name_title settled a launch readiness wait
on the shell's auto-title while Gemini was still booting. A launch now asks quiet
of every weak idle verdict, as that lane did.

Main's readiness census requires a recorder for every runtime fixture; the zsh
prompt recording is a non-agent control. Gemini's synthetic baseline is
regenerated for this PR's stated change: a bare gemini title is no longer its
rest mark, so name-only rows settle weak, and a fresh working or blocked status
is no longer overridden.

* fix(agent-launch): paste a launch prompt only when the launched agent is proven in front

A launch pasted its prompt unless a shell was proven in the terminal's
foreground, so any read that could not prove one let the prompt through. After
an agent exited at startup, its shell turned bracketed paste on at the next
prompt, readiness fired on it, and the prompt was typed into the shell:

- macOS: a pane runs its shell under login, so the process-group fence's root
  was never the shell's group and never proved it; the cached foreground name
  could also still name the exited process.
- Windows Git Bash and WSL: the shell-alone-in-its-job check never answers.

Now one fresh read of the terminal's foreground decides: agent, shell or
unknown. Only 'agent' lets a write through (paste, Enter, second Enter, reused
panes too); 'shell' still drops a ready signal. A Windows host never proves the
agent, so there the launch line carries the prompt at any size, as on main.

* test(agent-launch): cover the Windows QA stub, a grok override that exits at once

* fix(agent-launch): keep the local socket alive while a prompted launch waits for its agent

A launch with a prompt now waits up to 60 s for the terminal agent to be
ready before it writes the prompt, and reports not-delivered when the agent
never is. The local runtime socket closes a connection idle for 30 s unless
the request is a long poll, so a launch whose agent exited at startup lost its
reply and the caller saw 'runtime closed the connection' instead of
not-delivered. Classify a prompted agent.launch and agent.launchReplay as a
long poll, as orchestration.workerStart already is for the same wait.

* refactor(agent-launch): narrow the launch params by 'in' instead of a cast

* fix(agent-launch): find a launched agent behind a wrapper that leads its process group

A tcsh or nu launch line runs the agent from /bin/sh '<script>', and a
wrapper script that does not exec its agent does the same: the wrapper leads
the terminal's foreground process group and the agent is a member of it. The
fresh foreground read names the group's leader, sh, so a prompted launch was
refused or pasted late (M4Air tcsh: 2 of 4 not delivered, 2 pasted ~9 s late).

Before that read, take the host's process-group observation as positive proof
when it names the launched agent among the foreground group's members and is
younger than a ready signal's quiet window. It never proves a shell.

* fix(agent-launch): judge the foreground-group proof by when its capture began, not how long ps took

The age the host stamps on a process-group observation runs from the start
of its whole-machine ps, so on a loaded Mac a capture begun after the read
was asked for still read as older than 1 s and the proof was dropped. Count
an observation whose capture began after the read was asked for, less the
window a shared capture is reused across.

* test(agent-launch): keep the crash-guard live test out of the Windows lane's gate scan

The Windows-lane registration scan read the const assigned from a platform
check as a Windows-only gate, though the suite runs everywhere but Windows;
find zsh in a function instead, as the real-zsh typed-line test does.

Under load the fresh foreground scan can fail to answer, which lets the
shell's prompt settle readiness (2 of 4 paired runs). The guard still refuses
that write, so assert the refused write, the property that must always hold.

* perf(agent-launch): read a local pane's foreground from its own terminal, not the whole process table

The foreground read that gates every launch paste ran the daemon's
inspectProcess capture and then a fresh scan, each a whole-machine ps; the
fresh one also waits for any capture already running before it starts its own.
Measured here at load 5: 1.2 s a read (M4Air QA: 3.4-5.0 s, and worker starts
17.6-32 s against main's 9-12 s at load 25-84).

On a local macOS or Linux host, take the pane's root pid from the provider's
session inventory and run one ps limited to that pane's terminal. Its
foreground process group decides: the launched agent or any non-shell member
is the agent (a wrapper that did not exec its agent leads the group), a group
of shells alone is the shell. Same pane, same verdict: 2.7 ms a read. SSH hosts
keep the relay's observation and name.

* test(mobile): re-measure the web app's script sweep after agent.launch's capabilities moved out of protocol-version

The mobile web bundle check failed at 124 assets against a ceiling of 123. Main already sat
exactly on that ceiling: its sweep table read 69 scripts at 16 routes while the tree builds 73,
the whole margin of 4. This branch imports the agent.launch capabilities from their own module,
so protocol-version is no longer pulled into the root layout and four other routes. That moves
which routes share which modules, and the Qoder capability module, imported by protocol-version
and the AI-vault resume path, no longer shares an importer set with anything, so it gets a chunk
of its own: 74 scripts.

The fence says to re-derive the bound rather than raise it, so the sweep is re-measured on this
head (every prefix of the sorted route list). The worst route now adds 10 scripts (session), not
9, which moves the pinned shell crossing from 32 to 30 routes; main re-measured on its own lands
on the same crossing.

* fix(agent-launch): a worker's brief needs its agent found in front, and Grok's start answers on its composer

A paired-server worker start whose agent exited at startup typed its brief into the server's
shell, which ran it: the idle wait can settle on a shell back at its prompt, and the brief was
written with no foreground read. Both worker-start paths now check before each brief write, as a
launch prompt is checked: on a host that can find the agent in front it must be there; on one that
cannot (Windows) a shell proven in front still refuses, and anything else writes as before.

A Grok worker start waited ~10 s more than main: its only rest signal is its bare name, which a
launch holds to quiet output, and Grok animates its logo for ten seconds after its composer glyph.
A worker start for an agent whose rest signal is its bare name and whose composer draws a marker
(Grok, DSH, mimo-code) now also answers on that marker, whichever comes first.
2026-10-05 11:26:52 -07:00

311 lines
11 KiB
TypeScript

import { getDeepSeekBuildTitleStatus } from './dsb-terminal-title'
import { qoderTitleStatus } from './qoder-terminal-title'
import {
AGY_AGENT_NAME_RE,
BRAILLE_SPINNER_RE,
CLAUDE_IDLE,
CURSOR_NATIVE_TITLE_LOWER,
DROID_AGENT_NAME_RE,
GEMINI_IDLE,
GEMINI_PERMISSION,
GEMINI_SILENT_WORKING,
GEMINI_WORKING,
HERMES_AGENT_NAME_RE,
QUARTER_CIRCLE_SPINNER_RE,
STRONG_IDLE_KEYWORDS_RE,
STRONG_WORKING_KEYWORDS_RE,
STRONG_WORKING_KEYWORDS_RE_GLOBAL,
containsAgentName,
containsAgentSpinnerGlyph,
containsAny,
containsQuarterCircleSpinner,
containsLegacyAgentName,
isClaudeManagementTitle,
isDshTerminalTitle,
isGeminiTerminalTitle,
isPiAgentTitle,
isPiTerminalTitle
} from './agent-title-core'
import type { AgentStatus } from './agent-title-core'
import { isOpenCodeNativeTitle } from './opencode-terminal-title'
import {
getPiCompatibleTitleSeparatorStatus,
getPiCompatibleSyntheticAgentStatus
} from './pi-compatible-synthetic-title'
import { clearPiStateWorkingMarker, getPiStateTitleStatus } from './pi-state-title-marker'
import { getWrapperTitleSegments } from './terminal-title-wrapper-segments'
import { isGrokRotatingWorkingTitle } from './terminal-title-agent-type'
import { memoizeTitleClassification } from './terminal-title-classification-memo'
/**
* Strip working-status indicators so stale exit titles stop reporting working.
*/
export function clearWorkingIndicators(title: string): string {
// Why: Pi/OMP's static working marker survives every strip below, so a stale native
// title would keep re-arming the 3s clear timer without ever leaving working (#13890).
const clearedPiStateMarker = clearPiStateWorkingMarker(title)
if (clearedPiStateMarker) {
return clearedPiStateMarker
}
let cleaned = title
cleaned = cleaned.replace(GEMINI_WORKING, '')
cleaned = cleaned.replace(GEMINI_SILENT_WORKING, '')
cleaned = cleaned.replace(BRAILLE_SPINNER_RE, '')
cleaned = cleaned.replace(QUARTER_CIRCLE_SPINNER_RE, '')
if (cleaned.startsWith('. ')) {
cleaned = cleaned.slice(2)
}
if (containsAgentName(cleaned)) {
cleaned = cleaned.replace(STRONG_WORKING_KEYWORDS_RE_GLOBAL, '')
}
cleaned = cleaned.replace(/\s{2,}/g, ' ').trim()
return cleaned || title
}
/**
* Tracks agent status transitions from terminal title changes.
*/
export function createAgentStatusTracker(
onBecameIdle: (title: string) => void,
onBecameWorking?: () => void,
onAgentExited?: () => void,
initialTitle?: string
): {
handleTitle: (title: string) => void
seedTitle: (title: string) => void
restoreLastExit: (confirmedStatus?: AgentStatus) => AgentStatus | null
reset: () => void
} {
// Why: trackers restored mid-session need a last-known status without firing
// callbacks, or a hidden working agent can miss its later idle transition.
let lastStatus: AgentStatus | null =
initialTitle !== undefined ? detectAgentStatusFromTitle(initialTitle) : null
let restorableExitStatus: AgentStatus | null = null
return {
handleTitle(title: string): void {
const newStatus = detectAgentStatusFromTitle(title)
if (newStatus !== null) {
restorableExitStatus = null
}
if (lastStatus === 'working' && newStatus !== null && newStatus !== 'working') {
onBecameIdle(title)
}
if (lastStatus !== 'working' && newStatus === 'working') {
onBecameWorking?.()
}
// Why: reverting to a plain shell prompt after idle/permission means the
// agent exited; while working it can just be a transient internal title.
if (lastStatus !== null && lastStatus !== 'working' && newStatus === null) {
restorableExitStatus = lastStatus
lastStatus = null
onAgentExited?.()
}
if (newStatus !== null) {
lastStatus = newStatus
}
},
seedTitle(title: string): void {
lastStatus = detectAgentStatusFromTitle(title)
restorableExitStatus = null
},
restoreLastExit(confirmedStatus?: AgentStatus): AgentStatus | null {
const restoredStatus = confirmedStatus ?? (lastStatus === null ? restorableExitStatus : null)
if (restoredStatus !== null) {
lastStatus = restoredStatus
}
restorableExitStatus = null
return restoredStatus
},
reset(): void {
lastStatus = null
restorableExitStatus = null
}
}
}
/**
* Normalize high-churn agent titles into stable display labels before storage.
*/
export function normalizeTerminalTitle(title: string): string {
if (getDeepSeekBuildTitleStatus(title)) {
return title
}
const qoderStatus = qoderTitleStatus(title)
if (qoderStatus) {
const label = title.includes('Qoder CLI CN') ? 'Qoder CLI CN' : 'Qoder CLI'
return `${qoderStatus === 'working' ? '✦' : qoderStatus === 'permission' ? '▲' : '◇'} ${label}`
}
if (!title) {
return title
}
// Why: a Pi/OMP label is cwd/session text that may contain Gemini's glyphs; its own
// state marker is explicit, so it outranks glyph sniffing here as it does in detection.
if (!getPiStateTitleStatus(title) && isGeminiTerminalTitle(title)) {
const status = detectAgentStatusFromTitle(title)
if (status === 'permission') {
return `${GEMINI_PERMISSION} Gemini CLI`
}
if (status === 'working') {
return `${GEMINI_WORKING} Gemini CLI`
}
// Why only with the glyph: a bare `gemini` title (a shell auto-title) reads idle by default,
// and stamping Gemini's rest glyph on it turned a name into explicit readiness.
if (status === 'idle' && title.includes(GEMINI_IDLE)) {
return `${GEMINI_IDLE} Gemini CLI`
}
}
// Why: Pi/OMP animate a braille frame every 80ms, so the frame is the churn — but the rest of
// the title is the session name and cwd the agent chose. Canonicalize the frame in place
// (it leads in `⠋ π - session - cwd` and sits medially in `π ⠋ label`) and keep everything
// else; collapsing to a bare "Pi" discarded both the identity and the label (#16093).
// Why segments: a multiplexer prefixes the pane title (`zsh | ⠋ π - …`), and an anchored
// match would skip the canonicalization and let the frame churn through (#8032).
if (getWrapperTitleSegments(title).some(isPiAgentTitle)) {
return canonicalizeBrailleSpinnerFrame(title)
}
// Why: Grok Build interpolates a rotating status/tool phrase between the
// spinner and its name, so its working frames change the title many times per
// turn. Collapse them to one stable label; idle/session titles carry no
// spinner and pass through, so the meaningful final title still shows (#7863).
if (isGrokRotatingWorkingTitle(title)) {
return '\u280b Grok'
}
return title
}
/** Why: any braille frame reads as the same animation step, so consecutive frames dedupe. */
function canonicalizeBrailleSpinnerFrame(title: string): string {
let canonical = ''
for (const char of title) {
const codePoint = char.codePointAt(0)
canonical +=
codePoint !== undefined && codePoint >= 0x2800 && codePoint <= 0x28ff ? '\u280b' : char
}
return canonical
}
function computeAgentStatusFromTitle(title: string): AgentStatus | null {
const buildStatus = getDeepSeekBuildTitleStatus(title)
if (buildStatus) {
return buildStatus
}
const qoderStatus = qoderTitleStatus(title)
if (qoderStatus) {
return qoderStatus
}
if (!title || isClaudeManagementTitle(title)) {
return null
}
if (title.trim().toLowerCase() === CURSOR_NATIVE_TITLE_LOWER) {
return null
}
if (isOpenCodeNativeTitle(title)) {
return containsAgentSpinnerGlyph(title) ? 'working' : 'idle'
}
// Why: Pi/OMP's marker is an explicit state protocol, so it wins over the glyph and
// keyword gates below — its label is free-form cwd/session text that can carry either.
const piStateStatus = getPiStateTitleStatus(title)
if (piStateStatus) {
return piStateStatus
}
// Why the guard: DSH-TUI prefixes its title with `✦` while it is at REST, and that is
// Gemini's WORKING glyph. Reading it here made a finished DSH pane report working
// forever. DSH's own spinner prefixes are braille, which the spinner check below
// already covers, and its hooks are the authority either way.
if (!isDshTerminalTitle(title)) {
if (title.includes(GEMINI_PERMISSION)) {
return 'permission'
}
if (title.includes(GEMINI_WORKING) || title.includes(GEMINI_SILENT_WORKING)) {
return 'working'
}
if (title.includes(GEMINI_IDLE)) {
return 'idle'
}
}
// Why: resolve synthetic Pi/OMP permission/idle labels before the broader
// Pi and braille-spinner checks below.
const piCompatibleSyntheticAgentStatus = getPiCompatibleSyntheticAgentStatus(title)
if (piCompatibleSyntheticAgentStatus) {
return piCompatibleSyntheticAgentStatus
}
if (title.startsWith(`${CLAUDE_IDLE} `) || title === CLAUDE_IDLE) {
return 'idle'
}
// Why: read the state separator before the blanket idle below — `π ! <label>` is a
// blocked agent, and treating it as idle hides an OMP pane waiting on the user.
const piCompatibleSeparatorStatus = getPiCompatibleTitleSeparatorStatus(title)
if (piCompatibleSeparatorStatus) {
return piCompatibleSeparatorStatus
}
if (isPiTerminalTitle(title)) {
return 'idle'
}
if (containsAgentSpinnerGlyph(title)) {
return 'working'
}
const hasDroidAgentName = DROID_AGENT_NAME_RE.test(title)
const hasHermesAgentName = HERMES_AGENT_NAME_RE.test(title)
const hasAgyAgentName = AGY_AGENT_NAME_RE.test(title)
const hasLegacyAgentName = containsLegacyAgentName(title)
if (!hasLegacyAgentName && !hasDroidAgentName && !hasHermesAgentName && !hasAgyAgentName) {
return null
}
if (containsAny(title, ['action required', 'permission', 'waiting'])) {
return 'permission'
}
// Why: boundary-aware regexes avoid cwd/path and substring false positives.
if (STRONG_IDLE_KEYWORDS_RE.test(title)) {
return 'idle'
}
if (STRONG_WORKING_KEYWORDS_RE.test(title)) {
return 'working'
}
if (title.startsWith('. ')) {
return 'working'
}
if (title.startsWith('* ')) {
return 'idle'
}
// Why: Droid hook events are authoritative; native name-only titles should
// not turn a still-sleeping execute tool into completion.
if (hasDroidAgentName && !hasLegacyAgentName) {
return null
}
return 'idle'
}
/**
* Pure in `title`, so it is memoized on the title string: sidebar/tab selectors
* re-ask for the same unchanged titles on every store write.
*/
export const detectAgentStatusFromTitle: (title: string) => AgentStatus | null =
memoizeTitleClassification(computeAgentStatusFromTitle)
/**
* True when a quarter-circle spinner frame is the only agent evidence a title carries.
* Any TUI animates those glyphs, so they prove activity, not identity — callers that
* authorize writes into the pane need independent evidence (STA-4028, regression #13925).
*/
export function isQuarterCircleSpinnerOnlyAgentTitle(title: string | null | undefined): boolean {
if (!title || !containsQuarterCircleSpinner(title)) {
return false
}
return detectAgentStatusFromTitle(title.replace(QUARTER_CIRCLE_SPINNER_RE, '').trim()) === null
}