GitHub Copilot CLI (`copilot`) becomes a run mode: Run menu, welcome
screen, cron and CLI management once it is installed, tmux-backed like
the other agent CLIs, working/idle and the model read off its footer,
launch options through `copilotConfig` (model, resume, continue, the
multi-user-clamped allowAll), Custom Model Endpoints via Copilot's BYOK
variables, MCP sync of its mcp-config.json, a status probe, and the
`codeman agent` verbs (spawn waits for its composer, read returns its
last answer).
The plan-usage quota row that earlier rounds carried was split out on
review and is not part of this merge.
Merge-time commits on top: the wider COPILOT_* clamp and the zh-CN
README rows (maintainer), the Docker sign-in docs, divider rows inside
an answer kept by `agent read`, route tests for the screen-read
last-response branch, and doc drift (Past Sessions, wheelForward JSDoc,
COPILOT_/OMP_ in the env prefix lists).
Verified before merge on Copilot CLI 1.0.98: injected keystrokes,
bracketed paste and Enter reach the TUI in tmux (github/copilot-cli#4180
does not reproduce), a shell tool call completes inside tmux (#4223 does
not reproduce), and a session driven through this branch answered over
send-and-wait.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
_probePaneWorking() tests the pattern against the whole pane, so an answer
that quoted the spinner row (an agent explaining this entry: it matches
`⣯ Generating...`) kept the session busy for as long as the text was on
screen. Anchored like gemini's braille branch; every agy capture draws the
spinner at column 0. The prose test now covers the quoted row.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
- docs/cli-registry.md said Session history resume goes through
copilotConfig.resumeSessionId / continueSession. Past Sessions sends only
`continueSession: true` (terminal-ui.js, the modeConfigKey table);
resumeSessionId is for API callers.
- The wheelForward JSDoc in cli-registry/types.ts read as if the value had
effect. Nothing reads it (it is in the DECLARED-FOR-LATER list); the
browser gate is _shouldForwardWheelToApp in terminal-ui.js with its own
per-CLI rules, which the comment now says.
- The env-prefix lists in the CLAUDE.md Security table and the
allowedEnvPrefixes() comment in schemas.ts stopped at DEEPSEEK_, missing
both OMP_ and the new COPILOT_; the Multi-CLI prefix rule missed OMP_.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The COPILOT_ prefix admits every variable Copilot reads, and the 1.0.98
bundle reads far more than `copilot help environment` lists. Only
COPILOT_HOME, COPILOT_ALLOW_ALL and the BYOK trio were privileged, so a
non-granted multi-user owner could still hand back what the allowAll
clamp removed:
- approval: COPILOT_PLAN_THEN_AUTOPILOT, COPILOT_ASSISTED_APPROVAL
- code with no approval row: COPILOT_CLI_DIST_DIR runs <dir>/index.js
in place of the CLI, the package caches choose the CLI build,
COPILOT_EDITOR, COPILOT_PROVIDER_API_KEY_COMMAND and
COPILOT_AHP_HOST_BIN are executed, skills, provider configs and hooks
are loaded from the named places (the DSH_HOME class)
- token redirects: COPILOT_API_URL and the other API bases receive the
CLI's sign-in token, which in multi-user mode is the operator's (all
users share the OS account and ~/.copilot), and the rest of the BYOK
provider family re-aims a Codeman custom endpoint (the
DEEPSEEK_BASE_URL / OMP_AUTH_BROKER_URL class)
All are now privilegedEnvKeys. The registry schema capped that list at 8;
the cap is now 48, since a longer deny list only takes more away from
non-granted owners. The test sends one of each class through the real
schema (they all pass it) and the real clamp. The docs say the list is
read from the 1.0.98 bundle and that a variable a later release adds is
not covered until it is listed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The entry inherited wheelForward never from agentDefaults() while _shouldForwardWheelToApp forwards Copilot's wheel whenever the CLI tracks the mouse, at any version. New always mode, set on copilot, pinned by a registry test.
Like resumeSessionId, model follows a flag, so --model --yolo made --yolo its own flag and bypassed the allowAll clamp. New model-plain token pattern, the same rule in the create schema, and golden and schema tests.
Review of 8659ba25..be987fbb:
- Usage: a Free account answers premium_interactions {entitlement: 0, remaining: 0,
percent_remaining: 0, unlimited: false, has_quota: false}, which read as 100% used. The parser now
returns null when has_quota is false or the entitlement is not positive, with that payload as a
fixture. Free meters chat (200) and completions (2000) too, so the comment no longer says they
are unlimited on every plan.
- agent read: rows were skipped until a `❯ <prompt>` row, so an answer taller than the pane
(prompt scrolled off the top) read as "". With no prompt row AND no banner / trust dialog on
screen, the rows above the composer are now the tail of the last answer. Text typed in the
composer (a prompt-shaped row between the two rules) is not a prompt row.
- Page keys: measured on 1.0.95 in tmux that Copilot pages on PageUp/PageDown (three PageUp moved
lines 109-140 to 010-042, PageDown returned) as well as on SGR wheel reports, so copilot joins
claude and opencode in _localScrollbackIsHollow(); the stale-tracking-flag fallback the comment
and docs describe is now true, for the main pane and a TerminalTile (same function), with tests.
- X-GitHub-Api-Version is 2022-11-28, a version GitHub supports (verified against the live
endpoint); readPlanUsageTelemetryEnabled() is checked before isCopilotAvailable() so a switched-off
chip skips the resolver probe too.
- Palette label for Copilot in panels-ui.js; stale comments in types/session.ts and stock.ts.
- Docs: Past Sessions continues with --continue (resumeSessionId is for API callers), nothing sets
continueSession on a respawn, and only a tab name the user set is passed as --name; the usage
wording for Free; the wheel and mouse-strip paragraphs in CLAUDE.md and
architecture-invariants; the changeset describes the whole feature.
Copilot had no composerReadyMark, so `codeman agent spawn --mode copilot` returned as soon as
the session existed, before the composer could take a prompt (and before folder trust was
answered). The mark is the composer's footer hint, not the glyph: the trust dialog draws
`❯ 1. Yes`, so `❯` would call a worker ready while it still asks for trust. Measured on 1.0.95:
the dialog footer reads `↑/↓ to navigate`, the composer's reads `/ commands · ? help`.
Copilot's TUI turns mouse tracking on itself. Under `strip-mux-only` those DECSETs reached
xterm, which then reported a plain drag to the TUI instead of selecting (only Shift+drag
selected), so Auto Copy and copy-on-select did nothing. This is the failure opencode had and
`strip-mux-and-mouse` was added for: the DECSETs are stripped, a drag is a local selection and
clicks still reach the CLI through the browser's hand-encoded tap (`cliMouseTracking`).
Stripping them also stops xterm from encoding the wheel for the CLI, so
`_shouldForwardWheelToApp` now forwards it as SGR reports while Copilot is tracking the mouse
(the same bytes xterm sent natively, which is what scrolling was verified against). Shift+wheel
stays on local scrollback and a stale flag falls back to PageUp/PageDown.
Tests: the strip flavor for copilot (stripped DECSETs, 3J kept, cliMouseTracking, direct-PTY
untouched) and the wheel gate (forwards while tracking, no other non-claude mode changes).
- Drop GH_TOKEN / GITHUB_TOKEN from env.allowedKeys. The allowlist is one global list with no
mode context, so they were settable on every session. COPILOT_GITHUB_TOKEN is already admitted
by the COPILOT_ prefix.
- Declare capabilities.mcpConfig (copilot-json, COPILOT_HOME relocation) so Settings does not
list Copilot as unsupported while the sync-only target also lists it. The sync-only row is
skipped while the run mode is enabled.
- Require a leading alphanumeric on the resume id (path-segment in the registry, matching regex in
the schema): --resume's value is optional, so `--resume --yolo` made --yolo its own flag and
bypassed the allowAll clamp. Golden cases added; "ids only" wording corrected.
- Remove the unread overlays.credStore line and document signing in inside a Docker container.
- Add the zh-CN "Run GitHub Copilot" string, pin the new declarations in tests and update the
MCP registry expectations.
Copilot (`copilot`, npm @github/copilot) as a stock CLI entry with the same launch options as the other agent CLIs: model (--model), resume (--resume), continue (--continue) and allow-all (--yolo, clamped off for non-granted multi-user owners). Adds a resolver and GET /api/copilot/status probe, the agent-image package, Run-menu/mobile/logo entries, skill docs and tests.
Review round: the footer hint is drawn only at rest, so nothing arms the idle check while the Run this command list is up. The stock.ts comment, wiki, changeset and test now say a pending approval shows as working until answered, and the test pins busy during the confirmation. Adds the antigravity copy of the reattached-pane settle test, and updates the promptGlyph comment and the fixture header.
agy animates an 8-dot braille spinner (⣯ ⡿ ⣻ ⣾ ⣷ ⣟ ⢿ ⣽) with a label one row above
the composer, none of which is in SPINNER_PATTERN, and the entry declared no
workDetect, so no frame ever marked a tab working.
Declare the working line from the live pane and the footer hint `? for shortcuts`
as the idle-arming mark. The composer's `>` cannot be the glyph: the echoed
prompt starts with it too. Tests use verbatim captures from agy 1.3.3.
feat(codex): raise codex approval, question and trust dialogs as BLOCKED
Conflict resolved against #634 in src/session.ts _probePaneWorking():
#624's assignment (a dialog on screen is never work) is kept, and #634's
trust-dialog verdict block (log the hold once, remember the reading)
follows it, before _readWatching().
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
fix(deepseek): find pnpm when installing the terminal profile, and install it with dsh
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
On codex-cli 0.162.0, with 0.163.0 released, every launch first shows the
"Update available" prompt, and choosing Skip leads to a variant of the
workspace trust prompt: "2. Back to Agent Command Center" over the key-hint
row "enter continue · esc back". The pattern only accepted cancel,
interrupt, quit, answer or all as the closing word, so a fresh git repo
launched from Codeman sat on that trust prompt reading idle, with no red
alert and no blocked signal (measured live on a throwaway instance).
Adding "back" is safe because the hint row must still START with one of
the anchored hints ("enter continue" here): the popups the user opens at
the keyboard (/model, /permissions, /review, Implement this plan?) end in
"esc back" too, but their row starts with "enter select", so they stay
unmatched, as every existing negative fixture confirms. The new fixture is
the live capture with the scratch path replaced.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The DeepSeek install command asked npm for dsh and pnpm in one go. On a
machine that already has pnpm through corepack (`corepack enable` puts a
pnpm shim in node's own bin directory, which is npm's global bin for nvm
and Homebrew installs), `npm install -g pnpm` refuses to overwrite that
shim with EEXIST, and because both packages were in the same command the
whole install failed: dsh was not installed either. Measured with npm 10
against a scratch prefix holding a corepack-style shim. On master the same
machine installed dsh fine and the profile install used the existing pnpm.
The command is now `npm install -g @deepseek-ai/dsh && (command -v pnpm
>/dev/null || npm install -g pnpm)`, so pnpm is installed only when the
install environment has none, and a failed dsh install still fails the
command. The two hints that embed it follow (the launch-table drift test
requires that), and the catalogue is regenerated; install.sh withholds
the command for a launcher-only CLI, so its block is unchanged.
The test now runs the real stock string through /bin/sh with a fake npm
on a PATH it controls, for the three cases: no pnpm, an existing pnpm,
and a failing dsh install.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Codex is the one CLI Codeman launches with its approvals on by default, and
it declares `hooks: 'none'`, so nothing tells Codeman it stopped to ask. A
codex session parked on "Would you like to run the following command?" read
as plainly idle, and one parked on a question read as WORKING, because the
question's own key hints end in `esc to interrupt`, codex's working line.
This adds a generic, optional registry field,
`capabilities.workDetect.blockedLine` (plus `blockedLines`, the window), and
the session state that reads it:
- `isBlockedScreen()` (session-activity.ts, pure) joins the last
`blockedLines` non-blank rows of a capture and tests the pattern once, so a
pattern can demand several markers and anchor the dialog's footer on the
end of the window.
- `Session` reads it off the same capture the idle confirmation already
takes, emits `blockedChanged`, and exposes `screenBlocked`. It raises on
one reading, and while a dialog is up re-reads the screen every 2 s,
because answering resumes the turn and a busy pane would not reach the
settle path again until the turn ended. It clears only after two readable
captures in a row without the dialog, so a capture taken mid-repaint cannot
clear and re-raise it; an unreadable capture changes nothing. A blocked
screen wins over the working line in the same capture.
- The pattern goes through `compileVersionRegex()` in both the schema refine
and at use (lazily cached), like every other config regex.
Codex's pattern was measured on a live codex-cli 0.162.0 pane on 2026-10-14
(`-a on-request -s read-only`, widths 200 down to 36): it needs a `› N.`
option row and then the dialog's key-hint row as the last row, allowing for
that row to wrap below about 60 columns. Command, edit and network-escalation
approvals, single and multi-question dialogs and the startup trust prompt
match; the idle and working views, menus the user opened (`/model`,
`/permissions`, `/review`), the composer holding the dialogs' words and a
transcript quoting them verbatim do not. `Implement this plan?` is left out
on purpose: its footer is the same as `/model`'s and the turn has ended
there. The captures are kept as fixtures (trailing whitespace trimmed, the
scratch path replaced).
Nothing consumes `blockedChanged` yet; the next commit raises the alert.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Three comments said `altPath` is never written, but run() makes the alt file
the write target when the primary is absent. The behaviour is fine (a JSONC
file with comments fails the parse guard and is never written); the
comments now say so, so nobody loosens the path rule on that belief.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Conflicts resolved against #605 (and #602's test mock):
- CLAUDE.md Prompt uploads: #606's rewrite is kept and #605's bounded
workspace probe sentence follows the remote (SSH) refusal.
- test/routes/session-routes.test.ts: the remote-pane-history and
bounded-path-probe mocks are kept beside #606's _resetPasteRateBuckets
import, and both PRs' paste-image tests are kept.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Conflict resolved in terminal-tile.js's @dependency header: #601's
SHELL_LOAD_BASE_QUERY/SHELL_LOAD_QUERY and #527's pageKeysForGesture and
app.tabAlerts are all listed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The agent skill only reached Claude Code: injection wrote
.claude/skills/codeman, which no other CLI reads, and the per-session
preamble was seeded for claude alone. A Codex, Gemini, OpenCode,
Antigravity, Grok, DeepSeek or OMP session inside Codeman never learned it
could start and drive workers.
- Registry field capabilities.agentSkillDir (a closed enum, since
clis.json can set it and it decides where the server writes) names the
skills directory each CLI scans. Claude keeps .claude/skills; those seven
CLIs get the same skill in <case>/.agents/skills/codeman (each measured to
scan it; Claude Code does not, so it never sees two copies). pi gets none:
a project skill makes pi ask "Trust project folder?" at startup.
- The preamble is seeded for every local session whose CLI takes the
skill; nothing in it is claude-specific.
- The skill text is CLI-neutral and names the two CLI-side failures to
report instead of retrying: a sandbox that blocks loopback (Codex's
default), and commands that do not carry this session's environment
(Codex 0.162 runs them in a shared app-server daemon).
- POST /api/sessions injects after the launch pre-flight, so a CLI that
cannot start leaves nothing in the workspace.
- codeman skill install gains --mode; uninstall --case removes the copy
from every directory the registry can write.
- The settings toggle moves to Agents & CLIs > All agent CLIs.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Conflicts resolved by keeping both sides:
- terminal-ui.js: master's wheelDeltaWholeLines/sgrWheelReports next to
this branch's pageKeyRows/pageKeysForGesture.
- terminal-tile.js: master's tile SGR wheel forwarding and Shift+wheel
local scroll (_shiftScrollPending) kept; tile paging stays on the
shared pageKeysForGesture state (_pageKeyGesture) with its per-tile
dialog check.
- architecture-invariants.md: this branch's hollow-paging paragraph,
with master's tile SGR forwarding and Shift+wheel sentences replacing
the old "do not forward SGR wheel" note.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A phone-only toolbar Attach button hands the agent a photo or a video with the
keyboard down. It owns its own file input, positioned over the button so iOS
anchors the picker popover to it; the picked paths land on the prompt as unsent
text, so Compose adopts them together with anything already typed and Use
terminal keyboard hands a single-line draft back the other way.
Videos take a second path on the same route: ?kind=video streams the file to
disk under MAX_PASTE_VIDEO_BYTES (1GB, env CODEMAN_MAX_PASTE_VIDEO_BYTES)
instead of buffering it, sniffs the container from the first bytes as they
arrive, drains a mislabeled body unwritten and unlinks on any failure. It draws
on its own 3/min per-(IP, session) rate bucket beside the image path's 30/min:
at 1GB each, 30/min would be 30GB of disk a minute. Without the flag the route
is byte-identical to the image-only version, same 50MB cap and same bucket.
One client classifier, _promptAttachKind (MIME first, then the name's
extension), drives every pick and paste surface: the Attach button, terminal
drop, the tile grid's drop, the Compose and paste-dialog pickers and the
clipboard handlers. A pick that lands while the tile grid owns the terminal
goes to the PTY rather than the parked main-terminal overlay.
Gemini v0.63.0 treats a bare url as streamable HTTP first, OpenCode V2 nests
servers under mcp.servers with disabled instead of enabled, and a config kept only
as opencode.jsonc was read as "no OpenCode config". Fix all three read gaps;
the write path is unchanged. Fixes#586.
Conflict in docs/wiki/Settings-Reference.md resolved by keeping #565's Apply
wording and adding Copilot (and COPILOT_HOME, which the bot's review found
missing from this list).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Review follow-up after the rebase onto 1.40.0, where tiles began paging a
hollow buffer through the shared CodemanTerminalInput helpers and opencode
joined the hollow check (#555).
- opencode declares transcriptPageKeys in stock.ts, so the paging list read
by both panes is claude, codex and opencode. Without it, reading the
capability instead of the old id check would have turned opencode paging
off again. The stock-set and render tests name it.
- The gesture rules (first trackpad-sized event pages at once and is owed
back, wheel notches accumulate, a pre-paid debt dies with its gesture,
pinch and mostly horizontal swipes are consumed without paging) move out
of the primary pane into the pure CodemanTerminalInput.pageKeysForGesture.
The primary pane and TerminalTile both call it with their own state, so a
Codex or opencode tile pages exactly like the main pane.
- The dialog check stays per pane: the primary pane reads tabAlerts for the
active session, a tile for its own sessionId, so a wheel over a tile never
pages into that session's permission prompt while the focused tab is clear,
and a tile is not blocked by a dialog in another tab.
Tests: pure helper cases in terminal-scroll-routing, tile cases for Codex
paging, first-event paging, notch accumulation, pinch and swipe, byte parity
with the primary pane over a mixed sequence, and the per-session dialog
check in both directions.
Two changes to the last-resort page-key scroll (_maybePageCliTranscript), the
path a wheel or touch gesture takes when it is not forwarded to the CLI and
the local xterm buffer has no history (normal buffer, baseY 0).
Codex now takes that path too. Newer codex draws on tmux's alternate screen,
which the full strip hides from xterm, so a codex pane is as hollow as a
repaint-mode Claude pane and the wheel did nothing. Codex still ignores SGR
wheel reports and is still never forwarded the wheel; this sends plain
PageUp/PageDown keys, which codex does page its transcript on (measured on
0.157.1 locally and 0.160 over a remote session). Which modes may page is a
new registry capability, capabilities.transcriptPageKeys (claude and codex in
stock.ts), published to the page as window.__codemanTranscriptPageKeys, so
terminal-ui.js holds no CLI id for it.
The first event of a gesture now pages at once. The fallback used to send a
key only after half a screen of accumulated travel (19 rows, about 475px on a
38-row pane), so an ordinary trackpad flick sent nothing. The skipped travel is
owed back by the rest of the gesture, keeping one page per half screen; a pause
of more than 150 ms, a direction change or a tab switch starts a new gesture.
The agent skill teaches the session verbs to claude only (Codeman seeds its
preamble for local claude sessions). An opencode, codex, pi or gemini agent has
the same environment — CODEMAN_MUX, CODEMAN_SESSION_ID and CODEMAN_API_URL are
exported into every pane — and nothing that teaches it the verbs, so
`codeman agent ls|spawn|send|wait|read|interrupt|rm` packages them as commands.
It is a client of the server, like `codeman tui`, and stays out of
`codeman session` (which drives the in-process SessionManager). No new route and
no second transport: everything goes through CODEMAN_API_URL, so auth,
ownership and the per-session waiter cap apply unchanged. Credentials and the
Basic header come from src/codeman-credentials.ts, in the same order attach and
the TUI use.
Invariants, each in test/cli-agent.test.ts:
- Refuses outside a Codeman session (CODEMAN_MUX=1 + CODEMAN_API_URL); never
guesses a URL.
- `send` takes the prompt as ONE argument (an unquoted multi-line `$(…)` would
otherwise be split by the shell and re-joined into one line), transmits
printable text plus Enter only, and refuses multi-line input loudly instead of
letting sendInput weld the lines. No resend loop of its own: the server's
SubmitVerifier owns the swallowed-Enter case. ESC exists only as `interrupt`,
which never appends Enter.
- `rm` fails closed: empty id, an unprovable self id, or a prefix match in
either direction refuses.
- Nothing mode-shaped in the CLI: the readiness mark `spawn` waits for comes
from the registry (new `capabilities.composerReadyMark`: claude's composer
hint `shift+tab`, deepseek's `❯`), `read` relies on the route's own answer
dispatch, and a `stop`/`blocked` the session cannot fire is the server's 400,
passed through. A `/wait` timeout is a 200 with `timedOut`: exit 2 with a
neutral line, not an error. A worker that dies during spawn's readiness wait
is exit 3, like every other wait.
- `X-Codeman-Agent-Origin` rides only spawn's quick-start, the one request that
may create a case directory; every other verb leaves it off. Server-side,
test/routes/agent-case-marker-routes.test.ts pins that a POST /api/sessions on
an existing workingDir is never labelled, header or not, and that quick-start
labels only a directory it creates.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Copilot CLI keeps its user MCP list in ~/.copilot/mcp-config.json (COPILOT_HOME moves it) but is
not a Codeman run mode, so it has no registry entry. Add the copilot-json dialect (mcpServers,
tools ["*"], type local/http/sse, checked against `copilot mcp add` 1.0.94) and declare Copilot as
a sync-only target in src/mcp-sync-targets.ts, listed after the registry CLIs. A server switched
off with `copilot mcp disable` is recorded in settings.json (disabledMcpServers), not on the
entry: sync reads that list so it is not copied, and reports the target unreadable if the file
is not valid JSON instead of guessing.
- Codex footer model detection (c28): the modelDetect.screenLine effort
alternation is now built from CODEX_REASONING_EFFORTS plus 'default', so
'ultra' (offered by the codexReasoningEffort App Setting and codex's own
/model picker) is read and the launch enum and the footer reader cannot
drift again. Still one capture group, 125 characters, no new quantifier.
New session-display-model case loops every effort level, ultra included.
- No CLI-id branching for launch defaults (c27): the two mode === 'codex'
branches the synced codex model/effort defaults added to the create and
quick-start routes are replaced by a registry capability,
capabilities.launchDefaults (launch param -> settings key, values from a
closed enum), declared on the codex entry only. The resolver moved from
web/codex-launch-defaults.ts to web/launch-defaults.ts as
applyLaunchDefaults(mode, configs, customEndpoint), filling the entry's
legacyConfigField object through legacyConfigAliases, still re-validating
with SettingsUpdateSchema and never overwriting a caller's value. The
route exclusions are unchanged (create: not remote; quick-start: not
remote, not Docker, not a custom model endpoint), and quick-start still
derives the session model from a bag without ompConfig, as before.
schema.ts refuses an undeclared param, an unknown settings key, an empty
map, and launchDefaults on an entry with no legacyConfigField.
- The no-id-branching guard now carries an exact occurrence count per
allowlisted key, so a new copy of an already approved expression fails
instead of riding the old approval, with a synthetic anti-vacuity case.
- SettingsUpdateSchema JSDoc (c21/c29): 'classic' is the tabArrangement
default and 'compact' the headerStatsStyle default, matching the
resolvers and the pre-paint script; state/case/ledger are marked opt-in.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Comment and doc corrections that #555 made stale, no behaviour change.
- stock.ts: the grok and omp altScreen comments compared their strip to
opencode's, which is now strip-mux-and-mouse rather than the narrow
strip. Grok now says it shares antigravity's strip until measured, and
omp drops opencode from its comparison.
- terminal-ui.js: the touch-tap comment named Claude/Codex/Gemini as the
stripped modes, but the gate is now the cliMouseTracking flag alone and
covers opencode too, so it names the two stripping flavours instead.
- src/types/session.ts: the cliMouseTracking JSDoc (the flag the browser
now gates on exclusively) listed only claude/codex/gemini; it now names
the strip-full and strip-mux-and-mouse modes, including opencode under
tmux.
- src/session.ts: the usesMux getter doc now names isMuxMouseStripMode,
since the replay strip passes usesMux to it as well.
- docs/architecture-invariants.md: the narrow-strip list gains
grok/deepseek/omp (matching the PR's own CLAUDE.md line), the
"must REMEMBER" heading covers both DECSET-stripping flavours, and the
cliMouseTracking writer is described as the full-or-mouse branch it
really is.
- docs/wiki/The-Dashboard.md: the user manual said every non-Claude CLI
scrolls locally; opencode's wheel and swipes now page its conversation.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
An opencode tile showed only the session name, while Claude Code, codex and
DeepSeek tiles add `· <model>`: opencode declared no `modelDetect`, so its
screen was never read for a model and it is launched without a model param.
opencode draws the model on its composer's agent row, directly above the
box's bottom edge: `┃ Build Big Pickle OpenCode Zen`. Read from its own
1.3.0 source, the row is the agent, the model's name, the provider's name and
`· <variant>` when the model has one, and only colour tells model from
provider. So the field is all of it, exactly what opencode itself shows (the
owner's choice over a short id that only appears after the first reply).
- The pattern anchors on that row sitting directly above the `╹` edge, ends
the field at a double space (where the 200-column layout's sidebar shares
the row), skips the `No provider selected` placeholder, and takes the LAST
such row in the window through a lookahead, so a composer-shaped row the
agent prints higher up can never stand in for it. A test with a forged pair
inside the window fails without the lookahead.
- It reads 8 rows: the home screen puts up to five rows of opencode's own
chrome under the composer (key hints, a tip, the cwd/version row). The
schema's `screenLines` bound goes from 4 to 8, the reader's own cap; the
comment there records why a taller window is only safe with such a pattern.
- A permission prompt or shell mode hides the row; the last model is kept.
Measured against every captured opencode 1.3.0 frame (home screen and in
session, 40/60/120/200 columns, mid-turn and at rest, permission prompt):
the model was read everywhere it is drawn and nowhere else. Live on an
isolated instance from this branch, a restored opencode session published
`displayModel: Big Pickle OpenCode Zen` (source: screen) and its tile header
rendered `oc-home · Big Pickle OpenCode Zen`. The owner's own home-screen pane
on the 1.36.0 beta reads the same.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A gemini session stayed "working" for good after its first turn. Its braille
spinner trips the generic SPINNER_PATTERN and marks the pane working, but only
a composer glyph arms the idle confirmation and gemini declared none, so it
fell back to Claude's `❯`, which gemini never draws.
Measured on live Gemini CLI 0.63.0 panes (capture-pane every 300 ms through
real turns with a shell call at 40, 120 and 200 columns, YOLO and default
approval mode, plus the raw PTY stream). The turns ran against a local
stand-in for the Gemini API (GOOGLE_GEMINI_BASE_URL, which Codeman's custom
endpoint support already sets), since the CLI's TUI does not depend on the
backend and no account is needed for it:
- The TUI repaints its whole bottom region every frame, composer included,
and the composer sits between a `▄` bar and a `▀` bar; the submitted prompt
is echoed between the same bars. The `▀` bar arms the idle check: every
repaint carries it, tmux's reattach repaint too. The composer's prompt
character is no good: it follows the approval mode (`*` in YOLO), and its
`>` also starts the echoed prompt, which would make the submit verifier
read a submitted prompt as stranded and press Enter again.
- While a turn runs a line `⠦ Thinking... (esc to cancel, 6s)` animates about
every 80 ms (largest gap mid-turn: 214 ms). The label can be any loading
phrase, so the working line is the `(esc to cancel, <n>` suffix, or a
spinner frame opening a line for when a long phrase wraps that suffix.
Nothing at rest matches either.
- A tool confirmation (default mode) replaces the composer, stops the
spinner and the pane goes silent (3.9 s gap), so it reads as idle.
Verified on an isolated instance from this branch: a YOLO turn emitted one
session:working (+170 ms) and one session:idle (2.5 s after the last output);
a default-mode turn went working -> idle while the confirmation waited ->
working once allowed -> idle at the end; after a server restart four restored
gemini panes went busy -> idle in about 4 s; a fresh launch settled in 3 s.
The launch-settle and uncharacterised-CLI tests that used gemini as their
example of a CLI without work detection now use grok and deepseek.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A codex session waiting on a background terminal read as plainly idle,
with no "1 background terminal" badge. Codex pins that row above its
composer, and the registry looked for it in the last three non-blank
rows. Codex 0.162.0 added a hint row under the status line at rest
(` ← for agents · ? for shortcuts`), which pushes the chip to FOURTH from
the bottom exactly when the idle probe reads it. Measured live on the
1.36.0 beta with `sleep 600` started as a background terminal: chip,
composer, status line, hint row; the server reported `watching: null`.
While a prompt is typed the hint goes away and the chip is third again.
The codex entry now declares `watchingLines: 4`. The trade is stated in
the entry: with no terminal running, the fourth row from the bottom is
the last transcript row (the last two while typing), which the agent
writes. As before, this is contained by codex having no hooks: a forged
row costs a wrong badge, never a silenced alert.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A pi session showed no model in its tab or tile unless one was passed at
launch, and none at all for the default route. pi draws its model in its
own footer, but its registry entry declared no modelDetect, so the pane
probe never read it.
The footer, from pi 1.1.0's footer code (0.84.4's is the same) and a live
pane (`0.8%/253k (auto) qwen3.8-27b-pi • xhigh`): usage and context
on the left, then at least two spaces and `[(provider) ]<model>`, with
` • <thinking>` for a reasoning model and ` → <routed model>` when
routed. The last two rows are read (an extension status row can sit
below), and the context field picks the stats row out of them.
pi truncates the right side to fit a narrow pane with no ellipsis,
leaving exactly two spaces of padding. A name with nothing after it is
therefore read only with three or more spaces in front, and with two only
when a following ` •`/` →` proves it whole, so a cut-off name is never
shown. `no-model`, pi's placeholder, is rejected.
The read rides the idle confirmation pi gained with its workDetect entry.
Verified on an isolated instance: a fresh pi session published
qwen3.8-27b-pi (source screen) about 8 s after launch.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Every codex tile and tab showed no model, unlike claude and deepseek.
Codex reports its model only in the footer under the composer
(` GPT-6-Luna default · ~/codeman-cases/testcase`), and the registry read
the pane's LAST row for it. Codex 0.162.0 added a hint row under the
footer at rest (` ← for agents · ? for shortcuts`, or ` ? for
shortcuts`), so the last row was always the hint and `displayModel`
stayed null. Measured live on the 1.36.0 beta: the hint is there at rest
and after a turn, and gone while a prompt is being typed (the footer is
the last row again then).
The codex modelDetect window is now two rows, and the footer must be
either the last row or followed by exactly one more two-space-indented
row. Anchoring to the end of the window keeps the guard the one-row rule
had: with the footer hidden, the last two rows are a transcript line and
the `›` composer, so a footer-shaped line the agent printed is not read.
Tests cover both 0.162 hint variants, the typing layout, the hint never
read as a model, and a forged footer-plus-indented pair above the
composer.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
An omp session stayed "working" for good once a turn started, the same
latch pi had: omp's braille spinner trips the SPINNER_PATTERN fast path,
and only a composer glyph arms the idle confirmation. omp declared none,
so it fell back to Claude's `❯`, which omp never draws once its setup
wizard is done.
Measured on live omp 18.8.6 and 18.0.11 panes, holding a turn open
against a local endpoint that never answers: the input row is `╰─ <text>`
and is redrawn at submit, at the end of a turn, at launch and on
reattach. While a turn runs, the status bar's leading `π` becomes a
braille spinner plus the elapsed time (` ⠼ 14s > ⬢ model > ...`; 18.0.11
pads it with two spaces, past a minute it reads `1m`), with a
`⎋ Working…` row above it. The registry entry now names the input row as
the glyph and either working signal as the working line.
The glyph also switches the submit verifier on for omp, which reads the
input row the way it reads Claude's composer. A prompt sent mid-turn goes
to omp's Steering queue and clears the row, so the verifier stands down.
Text left in the row after an Enter is the one case it re-presses.
Verified end to end on a sandboxed instance (own HOME and PATH, omp
18.8.6): session:idle at launch, session:working during a turn,
session:idle about 3 s after it ended, and a restored pane settled idle
about 3 s after a server restart.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
An opencode session that ran a tool stayed "working" for good. The running
tool row draws a braille spinner (`⠋ Sleep for 12 seconds...`), which trips
the generic SPINNER_PATTERN and marks the pane working, but only a composer
glyph arms the idle confirmation and opencode declared none, so it fell back
to Claude's `❯`, which opencode never draws. Measured on an isolated
instance: a 16 s turn latched busy/isWorking for the rest of the session.
A text-only turn had the opposite problem and never showed as working.
Measured on live opencode 1.3.0 panes (capture-pane every 250-300 ms through
real turns at 40, 60, 120 and 200 columns, plus the raw PTY stream):
- Every composer row starts with a `┃` bar, and the submitted prompt lands in
the transcript with the same bar, so a turn's first repaint arms the idle
check, and tmux's reattach repaint does the same for a restored pane.
- While a turn runs the footer row starts with an 8-cell knight-rider
spinner, `⬝■■■■■■⬝ esc interrupt`, redrawn about every 40 ms (largest
gap mid-turn: 121 ms). At rest the TUI is silent and nothing on screen
draws a `⬝`/`■` run, the 200-column sidebar included.
- The working line is the spinner run, `[⬝■]{8}`, not the label: tmux ships
`esc` and `interrupt` as separate words joined by cursor moves, so the
label never reaches the stream detector, and at 40 columns the footer
wraps it to `esc` / `interr` / `upt`. All 344 spinner chunks of a turn
match the run after Codeman's ANSI strip.
- A pending permission prompt replaces the composer and stops the spinner,
so it reads as idle (waiting on the user).
- The last `┃` row on screen is the composer's agent/model row, or the
permission box's closing bar, never the prompt text, so the submit
verifier stands down and can never press Enter into a dialog.
Verified on an isolated instance from this branch: a 15 s tool turn emitted
exactly one session:working (+271 ms) and one session:idle (3 s after the
spinner stopped); a permission prompt read idle and the allowed turn went
working -> idle; after a server restart both restored opencode panes (one
at rest, one on a permission prompt) went busy -> idle in about 4 s; a
fresh launch reached an open page as idle in 3 s.
The launch-settle tests that used opencode as their example of a CLI
without work detection now use gemini and antigravity.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A pi session stayed "working" for good once a turn started. pi's braille
spinner trips the SPINNER_PATTERN fast path, which marks the pane working,
but only a composer glyph arms the idle confirmation and pi declared none,
so it fell back to Claude's `❯`, which pi never draws. Measured on beta136:
an errored turn stayed busy/isWorking for 3+ minutes after pi was back at
rest.
pi has no composer glyph. Measured on a live pi 1.1.0 pane (capture-pane
every 250 ms through a turn): its composer sits between two `─` rules, and
while a turn runs it rewrites the top rule as `── ⠏ Working ───` on every
frame. The registry entry now names the rule as the glyph that arms the
check and a spinner frame inside it as the working line.
The same glyph settles a reattached pi pane (a restored pane gets no launch
timer): tmux's reattach repaint carries `─`, which arms the confirmation.
The submit verifier reads the last rule, finds no prompt text and stands
down, so it can never press Enter on a pi pane.
Verified on an isolated instance from this branch: a real pi turn emitted
session:working then session:idle, and after a server restart the restored
pi pane went busy -> idle in about 3 s.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
opencode's TUI enables mouse tracking. tmux runs with `mouse off`, so it passes
the PANE's DECSETs straight through to the tmux client, and the browser's xterm
obeyed them: `mouseTrackingMode` flipped to 'any' (measured 62 none / 18 any over
16s) and xterm then reported DRAGS to the TUI instead of selecting locally.
In that state marking text produced no selection at all, so copy-on-select
silently did nothing (5/5 dead drags while `any`), and the obvious fallback —
Ctrl+C — is opencode's `app_exit`, which ended the session. Both were hit here.
opencode needs the middle strip: alt-screen toggles AND mouse DECSETs, but NOT
`3J` (a TUI is not a `clear` consumer). That is `altScreen: 'strip-mux-and-mouse'`
+ `isMuxMouseStripMode`, applied to the live stream (session.ts) and the replay
of a stored buffer, now the exported `stripReplayBuffer()` (session-routes.ts).
The browser's mouse-report gate keeps no mode list any more:
`_shouldReportMouseToCli()` reads only `cliMouseTracking`. The server sets that
flag solely in the mouse-strip branch (`_recordStrippedMouseMode`, one caller),
so it can only be true for a mode whose DECSETs are stripped, and whichever modes
the registry strips, the browser follows. Clicks still reach opencode through the
hand-encoded SGR tap it gates.
The `altScreen` JSDoc gets the decision table its three independent choices need
(alt-screen / `3J` / mouse DECSETs), written from the predicates, including that
`preserve` and `strip-mux-only` take the same runtime row. The table is pinned for
every stock CLI, with and without tmux, on both the live strip and the replay
strip, plus the published flag (test/claude-scrollback-strip.test.ts), so the two
halves cannot drift and a mis-ordered replay branch fails.
Docs and comments that still said opencode keeps its mouse reporting or gets the
narrow strip are updated (CLAUDE.md, architecture-invariants, scrollback and
copy-shortcut plans, session.ts, terminal-ui.js).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The digit rule from 21ae48a5 hid the official DeepSeek ids (`deepseek-chat`,
`deepseek-reasoner` carry no digit), so a session on the official route with
the model field on showed the logo alone (its bundle row pins a provider
alone, so the config had nothing either). It also still misread a folder name
with a digit when every field before it was off.
Now the captured field is rejected when it is what the field can be when it is
NOT the model, and read otherwise:
- capabilities.modelDetect.rejectWords (registry data, single tokens, compared
ignoring case; the schema bounds them and requires a screenLine). dsh lists
every effort id its adapters offer (pi-ai THINKING_LEVELS plus the DeepSeek
adapter's off/low/high/max) and the shipped mode ids, from dsh 0.1.1-rc.2 /
dsh-TUI 0.10.0-beta.1. A mode's drawn label (`plan mode`, `full access`,
CJK) can never be one captured field.
- In the shared screen reader, for every CLI: a field equal to the session's
own working-directory basename is the folder, never the model.
Fixtures: `deepseek-chat` and `deepseek-reasoner` with the model field on are
read; every effort id, `default`, `plan mode`, and the folder name first (with
and without a digit) are not; the live qwen footer still reads `qwen3.8-27b`.
Known gaps, all off by default, are named in stock.ts: a custom mode id drawn
raw, a git branch or a one-word session title first, and the non-compact
footer layout (nothing read there; the route config applies).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
dsh-TUI draws the model as its status line's first field only while the
status bar's model field is on (the default). Switched off, the first field
is the reasoning effort (` medium · th-scratch`), else the mode, else the
cwd's basename, and the footer pattern read that as the model, which would
also outrank the route config added in the previous commit.
The captured field must now carry a digit, as a model id does (a version) and
an effort word, a mode name or most folder names do not. A model id without
one (`deepseek-chat`) is not read from the screen and the session falls back
to its route config: silent, never wrong.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
displayModel gains a `config` source, ranked below any report from the running
CLI and above the launch model: custom endpoint, then statusline or screen,
then config, then launch, then nothing. The screen still wins whenever it
names a model, since that is what the running TUI uses.
- Registry data: capabilities.modelDetect gains `configResolver`, a NAMED
reader (src/model-config-resolvers.ts), like a launcher profile; dsh names
'deepseek-route' (the reader from the previous commit). `screenLine` becomes
optional; the schema refuses a modelDetect naming nothing, an unknown
reader, or screenLines without a screenLine.
- Session: the reader runs from _withPaneLifecycle's finally, so at every pane
start, attach and relaunch, with the session's own launch config
(legacyConfigForMode) and env (its clamped overrides, then the server's), so
a per-session DSH_HOME is the home read. Async; a read that lands after a
newer one or after the session stopped is dropped; a remote or docker
session reads nothing locally. A change emits displayModelChanged
(broadcast and persist). Not restored after a restart: the next attach
reads it again, and a restored screen value outranks it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A session header can only name the model a session runs if the server knows
it, so SessionState gains `displayModel: { model, source }`, resolved in a pure
module (src/session-display-model.ts), strongest first:
- custom-endpoint: a Custom Model Endpoint Profile's modelId answers the
session, whatever alias the CLI prints;
- statusline / screen: the newest report from the running CLI itself.
Claude's statusLine exporter already posts model.display_name on every
render; the status-telemetry route now records it (only for a CLI with
capabilities.statusLineTelemetry). A CLI whose registry entry declares the
new capabilities.modelDetect has its footer read off the pane capture the
idle/working probe already takes (no extra tmux call), so an in-session
/model switch is followed at the next transition;
- launch: the model the session was launched with (claude's --model or the
app-wide default, another CLI's <cli>Config.model), read where the registry
says the model param lives;
- nothing known: no field, never a placeholder.
modelDetect is registry data, measured on live panes: dsh-TUI's status line
on the row under its composer (qwen3.8-27b on the owner's route) and codex's
`<model> <effort> ·` footer on its last row. Both anchor on chrome only that
CLI draws, over the last rows of the screen only; a transcript line shaped like
the footer is never taken (fixture tests). The pattern goes through
compileVersionRegex() with exactly one capture group, checked at load time.
An unreadable or covered footer keeps the last model (unlike the watching
label: a model does not stop running when something covers its row). Model
text is untrusted: escape sequences and control characters are stripped and it
is capped at 64 characters. A change emits displayModelChanged, broadcast
(session:updated) and persisted; a restart restores a CLI-reported model until
the next report.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
- The doctor now judges candidates like the run mode's resolver: the PATH
hit, then each search dir, each one version-checked on its own and
skipped on a mismatch (a wrong `pi`/`grok` on the PATH no longer hides
the real one in a search dir). A search-dir candidate must be an
absolute path to an executable regular file, so a relative dir or a
file without the x bit reads as missing, as it does in the Run menu.
`isExecutableRegularFile` is exported from cli-executable-resolver.ts
and reused rather than copied.
- Every doctor probe passes killSignal: 'SIGKILL'; a --version that
ignores SIGTERM held the probe for its full runtime (15 s vs 5 s
measured with a TERM-trapping script).
- README no longer claims parity with the Run menu or nvm prefixes.
- The Diagnostics panel marks a missing optional tool with ○, a missing
required one with ✗, as the terminal doctor does.
- expandSearchDir names its twin, expandHome() in cli-resolver.ts.
- test/doctor-cli-json.test.ts is hermetic: temp HOME, a PATH of only
`which` and `node`, and a clis.json that drops the registry's absolute
search dirs, so it never runs the machine's installed agent CLIs.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
fix(cases): bound path probes for linked workspaces and session creation, so an unreachable mount cannot freeze the server
# Conflicts:
# src/web/routes/case-routes.ts
Codeman maintainer