ensureBaseDirs unconditionally called chmod(baseDir, 0o700) on every startup
and on every team_create. On filesystems where the OS rejects chmod for the
directory (network mounts, SIP-protected locations, non-owner cases on macOS
shared by multiple GUI users), the call raises EPERM and the entire team-mode
init aborts:
[team-mode] init failed: EPERM: operation not permitted, chmod '/Users/<u>/.omo'
Wrap chmod through a small safeChmod helper that converts EPERM, ENOTSUP, and
EINVAL into a single warning log and continues. mkdir already creates new
directories with mode 0o700, and the existing post-creation stat-guard remains
in place for the case where the directory pre-exists with a different mode and
chmod is permitted, so the security envelope on supported filesystems is
unchanged. All other error codes (ENOENT, EACCES, etc.) still propagate.
Regression test mocks node:fs/promises.chmod to throw EPERM and asserts that
ensureBaseDirs completes successfully and emits exactly the documented warning.
Require the latest assistant tool-wait turn itself to be stale before a shouldReply parent wake can bypass tool-call deferral. This prevents an all-complete background wake from forking a second parent prompt loop when OpenCode has repaired the tail to a synthetic user message.
Tests:
- bun test src/features/background-agent/parent-wake-user-message-race.test.ts src/features/background-agent/task-completion-cleanup.test.ts src/hooks/shared/prompt-async-gate.test.ts src/shared/prompt-async-route-audit.test.ts --bail
- bun run typecheck
- bun test
After the 4.2.0 unified-dispatch refactor (a42f894f / df198d8b / fee515c5 / 989ab717 / dd3fecaf / 1bbe065c / 12bd6580), at least one caller in the new prompt-async-gate path forwards a FallbackModelObject (or some other non-string shape) into parsers that statically claim 'model: string'. The downstream .trim() call then throws 'model.trim is not a function', which rejects the session.processor promise and surfaces as 'Aborted process' + UI 'interrupted'. The issue (#4145) reports this aborts 90% of subagent dispatches across every provider on 4.2.0 + opencode 1.15.4.
This patch adds a 'typeof x !== "string"' runtime guard at the four parser entrypoints called from the dispatch path:
- src/shared/fallback-chain-from-models.ts :: parseVariantFromModel, parseFallbackModelEntry
- src/tools/delegate-task/model-string-parser.ts :: parseVariantFromModelID, parseModelString
- src/shared/model-string-parser.ts (duplicate file with same API) :: parseVariantFromModelID, parseModelString
- src/features/claude-code-agent-loader/claude-model-mapper.ts :: mapClaudeModelString
Each parser now returns undefined / { modelID: "" } for non-string input instead of throwing. This unblocks subagent dispatch and leaves the underlying caller bug for a follow-up.
Regression coverage: three new tests in src/shared/fallback-chain-from-models.test.ts pin the non-string behavior (object, null/undefined, number). Existing 38 tests still pass. Total: 41/41 green, typecheck clean.
Background fallback retry notifications were queued as bare internal user messages, so OpenCode could treat the notification as a new default-agent turn. Reuse the same parent prompt context resolver used by completion notifications for retrying and retry-ready wakes, and pin regression coverage for Hephaestus parent sessions plus missing-context fallbacks.
buildShellAwareGitPrefix incorrectly returned raw bash VAR=value prefix for csh, which does not support inline env assignment. Now routes csh through buildEnvPrefix() to emit setenv syntax. Also sets code block lang to csh and skips bash block regex prefixing for csh.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Use detectShellType() and buildEnvPrefix() from src/shared/shell-env.ts instead of hardcoding bash-only VAR=value syntax. PowerShell users get $env:VAR='value'; cmd users get set VAR="value" &&; unix/Git Bash users keep VAR=value. Skips injecting non-bash prefixes into bash code blocks to avoid syntax mismatch.
Reland of #3214 (ekkoitac) which had unresolvable CLA + used custom shell detection instead of the shared utility.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: ekkoitac <lobster@example.com>
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
When a background subagent emits [ALL BACKGROUND TASKS COMPLETE], the
plugin queues a parent-wake that ultimately calls
dispatchInternalPrompt against the parent session. If the user submits
a new prompt inside the ~250 ms post-dispatch hold window, both writes
land on the same OpenCode session-storage file at the same instant.
OpenCode's @parcel/watcher (which the plugin itself does not depend on,
but does indirectly trigger) batches those events into a TSFN callback
and dispatches them into a JS env that the renderer has just torn down
because the session view re-mounted around the user's new message ->
napi_fatal_error / SIGABRT on macOS arm64. Removing the plugin removes
the parent-wake, which is why removing OmO eliminates the crash.
Mitigation:
- Before flushPendingParentWake calls dispatchInternalPrompt, inspect
the parent session's message tail. If the most recent message is a
user message added inside PARENT_WAKE_USER_MESSAGE_IN_PROGRESS_WINDOW_MS
(default 2_000 ms), reschedule instead of dispatching. The user's own
prompt will drive the model; queued notifications will be re-flushed
on the next idle.
- Best-effort unref() of the long-lived pending-retry and dispatched-
wake bookkeeping setTimeouts. They previously pinned the host event
loop and prolonged the teardown window during which the watcher race
can fire.
The new option userMessageInProgressWindowMs is wired through
BackgroundManager via a module-level constant and is independently
testable.
Regression test parent-wake-user-message-race.test.ts covers:
- fresh user message -> dispatch deferred
- latest message is assistant -> dispatch proceeds
- user message older than window -> dispatch proceeds
- window=0 disables the guard
This is a surface-level mitigation of the most-likely root cause from
the audit; a deeper fix (singleton guard against plugin
double-instantiation under @opencode-ai/plugin@local reload, dispose
lifecycle for OpenCode plugin reload) is out of scope here.
Root cause of the user-visible `/init-deep ulw` hang (session
`ses_1cb9c3013ffesUOy5H3QOIya4K`): the plugin's `unhandledRejection` and
`uncaughtException` listeners were calling
`scheduleForcedExit(handler(error), 1, true)`, which both ran the entire
`cleanupAll()` chain (BackgroundManager shutdown, tmux pane closure,
team-mode teardown) and then `process.exit(1)`'d the host. Under heavy
slash commands like `/init-deep ultrafucking deep`, a single mid-stream
error (e.g. opencode's own `session.processor` Aborted-process condition,
or a transient socket reset) would:
1. trigger the listener,
2. abort the in-flight background tasks (`session.error
MessageAbortedError` for both child sessions in the log),
3. close the tmux panes the user was watching,
4. immediately kill the host via `process.exit(1)`.
From the user's seat that looked like a frozen TUI, which is what they
reported as "ulw 여전히 멈추는데". The error blob also logged as `{}`
because `JSON.stringify(new Error(...))` strips non-enumerable Error
fields, so the previous log line carried no diagnostic value.
This change makes the global `uncaughtException` / `unhandledRejection`
listeners log-only:
* New `describeProcessCleanupError()` extracts `{name, message, stack}`
from Error instances, falls back to a structured `{raw: ...}` payload
for plain objects / primitives, so the log now actually says what
failed.
* `registerErrorEvent()` no longer runs cleanup and no longer calls
`scheduleForcedExit`. It detaches itself, logs a single explanatory
line, and returns. Bun's default crash behaviour is already suppressed
for these events when a listener is present, so the host now genuinely
survives transient streaming errors instead of being killed by our
own helper.
* Signal handlers (`SIGINT` / `SIGTERM` / `SIGBREAK` / `beforeExit` /
`exit`) keep their existing behaviour and still run `cleanupAll()`
before the host terminates — that is now the only path that tears
down background tasks and tmux panes.
Tests are updated to lock in the new contract:
* New regression `#given scheduleForcedExit enabled AND unhandledRejection
fires #when the listener runs #then process.exit is NOT called AND
process.exitCode stays 0 AND no cleanup runs` (and the
uncaughtException twin) re-enables `scheduleForcedExit`, spies on
`process.exit` plus `globalThis.setTimeout`, and asserts none of them
are touched. Without the fix this test failed exactly like the
observed hang (exit called once, exitCode set to 1).
* The existing "manager shuts down before process exits" tests are
rewritten to assert the opposite: cleanup is NOT invoked from the
error path.
* A complementary `'exit'` listener test pins the real shutdown
contract (`exit` event still triggers `cleanupAll`).
* A new `#given describeProcessCleanupError` block covers the four
shapes (Error, plain object with own fields, empty object, primitive).
* The unregister assertion is tightened to check the listener count
drops back to baseline (previously it relied on a side effect of the
old cleanup-on-error path).
Manual QA: `bun /tmp/process-cleanup-smoke-test.ts` (out-of-tree smoke
driver) emits five back-to-back unhandledRejection/uncaughtException
events with various payload shapes and prints
`SMOKE_TEST_OK survived 5 emissions; exitCode=0; shutdownInvocations=0`,
confirming the host survives and no spurious cleanup runs.
`bun test` runs green for the affected modules:
- src/features/background-agent (528 tests)
- src/create-managers + src/plugin (218 tests)
- src/hooks/{unstable-agent-babysitter,ralph-loop,keyword-detector}
(225 tests)
External review on PR #4074 noted that adding setSessionAgent for child
sessions left sessionAgentMap holding entries after the session was
deleted or after the sync call_omo_agent executor cleaned up other
owned state. The map only grows; entries never get reused but they do
accumulate across long-running plugin instances.
Close both gaps:
- BackgroundManager.handleEvent for session.deleted now calls
clearSessionAgent for the deleted session id on both the early-return
no-task branch and the cascade tail. This pairs with the existing
clearDelegatedChildSessionBootstrap and SessionCategoryRegistry.remove
so all owned session state is dropped together.
- sync-executor finally for createdSessionForExecution now calls
clearSessionAgent alongside the existing subagentSessions,
syncSubagentSessions, and deleteSessionTools cleanup so sessions this
executor created cannot leak their agent mapping.
Adds focused tests:
- BackgroundManager.handleEvent - session.deleted cascade > should
clear session agent state for deleted sessions to prevent map leak
- executeSync > registers child-session bootstrap and tracked prompt
state before sync prompt dispatch (extended assertion for cleanup)
Two adjacent gaps cubic flagged on the previous diff:
1. spawner.startTask stored input.agent (potentially prefixed with sort
marker and ZWSP) in setSessionAgent, but the prompt body used the
stripped/normalized form. The session-agent registry therefore did
not match what promptAsync actually dispatched. Capture the
normalized agent once at the top of startTask and use it for
setSessionAgent plus the launch log lines.
2. manager.startTask wrote setSessionAgent(sessionID, input.agent)
before the cancelled and stale-attempt cleanup branches, but those
branches only cleared subagentSessions and the delegated bootstrap.
The session->agent mapping survived as orphan state after an aborted
launch. Call clearSessionAgent inside both early-return paths so
nothing remains tied to a session we just aborted.
Adds focused tests for both: spawner persistence parity with promptAsync
and manager cancellation cleanup leaving getSessionAgent undefined.
Three coupled gaps surfaced after the initial spawn fix:
1. fallback-retry-handler dropped task.skillContent and
task.sessionPermission when rebuilding LaunchInput, so the retried
background task lost the delegated system prompt and question-deny
permission rule.
2. manager.startTask never bound the child sessionID to the resolved
agent via setSessionAgent, leaving runtime fallback and other hooks
with no idea which agent owned the new child session.
3. The fallback-to-general path in spawner.ts rebuilt the prompt body
without going through buildFallbackBody, so bootstrap state, session
tools, and session agent updates drifted apart.
Persist skillContent and sessionPermission on BackgroundTask, bind
setSessionAgent/updateSessionAgent at session creation and on fallback,
and route the FALLBACK_AGENT retry through buildFallbackBody so the
prompt body, bootstrap tools, and session registries all agree.
- script/run-ci-tests.ts: CI test sharding and isolation logic
- script/run-ci-tests.test.ts: tests for CI test target selection
- src/features/background-agent/session-route.ts: session prompt routing for background agents
- src/hooks/interactive-bash-session/parser.ts: interactive bash output parser
- src/hooks/ralph-loop/completion-promise-detector-test-input.ts: test fixture for completion promise detection
Preserve delegated child prompt/bootstrap metadata for early runtime fallback before OpenCode has persisted the first user turn. Bind prompt gate calls to the SDK session receiver and keep completed background task lookup visible across plugin manager instances.
When a team-mode subagent hit a fallback model (rate limit / quota
exhaustion on the primary), the fallback continuation started a
fresh subagent session that was not registered in the team's
member registry under the original role. Subsequent
team_send_message / team_status calls from the fallback agent
threw "not in team" because the membership lookup missed.
Capture teamRunId + member identity at fallback initiation and
carry them onto the fallback session so the fallback agent
remains a first-class team participant. If preservation is not
possible, surface a bounded structured error instead of letting
the runtime fail mid-flight with a confusing membership message.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When oh-my-opencode created a delegated tmux pane, terminal
capability/color probe replies emitted by tmux or the freshly
attaching opencode session could end up in the caller pane's
input buffer instead of being consumed by the delegated pane,
appearing as literal text in the main OpenCode chat (e.g.
"414/21212a2/...").
Root cause: buildSplitArgs in team-layout-tmux/layout.ts called
split-window without the -d (detached/don't-switch-focus) flag.
Without -d, tmux briefly grants focus to the new pane during
creation; the outer terminal then sends DA1/DA2 and OSC color
probe replies into what it believes is the active pane, but the
focus handoff races and those bytes land in the caller pane's
stdin buffer instead.
Fix: add -d to every split-window call in buildSplitArgs, matching
the same flag already used in pane-spawn.ts for inline subagent
panes. This keeps the caller pane focused throughout the delegated
pane lifecycle so probe replies are consumed by the correct target.
Existing tests pass; one new test asserts -d is present on every
split-window call to guard this invariant.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When a hard-reject agent (e.g. prometheus) called team_create with an
explicit `lead` in the spec, the eligibility check that runs in the
no-lead branch of shouldReuseCallerLeadSession was bypassed. The
caller session was never registered in the team, the spawned lead
ran as a detached child, and replies routed to the spawned lead
never reached the caller — the caller became an orphan that could
send but never receive.
Move the caller eligibility guard to the top of team_create.execute
so it runs unconditionally before any team-run state mutates. Throw
an actionable error naming the agent and explaining hard-reject
agents cannot lead teams regardless of an explicit `lead` in the
spec.
When ctx.serverUrl had a port string of "0", TmuxSessionManager
silently replaced it with the localhost:4096 fallback and
createTeamLayout subsequently skipped pane creation without any
user-visible signal. The two-step silent failure made team_mode
tmux_visualization look broken in default TUI mode.
Surface the failure path:
- TmuxSessionManager now retains ctx.serverUrl on the instance and
exposes it via getCtxServerUrl(), and emits a structured warning
log on the port-0 fallback branch naming both the discarded URL
and the fallback it landed on.
- createTeamLayout's "opencode server not reachable" log is
upgraded to a structured warning including ctxServerUrl and a
hint to launch with --port N + OPENCODE_PORT=N.
No behavior change to the fallback resolution itself - only the
silence. Existing port-0 fallback tests still pass; two new tests
assert the warning fires on port 0 and is absent for real ports.
Replace the inlined parent-wake coalescing logic in manager.ts with delegation to the ParentWakeNotifier extracted in c1ccf8d09. The four timer Maps and the related methods now live in their own module with a narrow public API, while BackgroundManager retains the wiring point and the enqueue-callback bridge.
Closes HIGH-9 (step 2: integration)
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Extracts the parent-wake coalescing logic (pending/dispatched wake maps,
timers, notification reply assembly) from manager.ts into a standalone
ParentWakeNotifier class. Takes dependency-injected client, directory,
and an enqueueNotificationForParent callback, so the manager can delegate
parent-wake state to a narrow API.
This commit only introduces the new module; wiring manager.ts to use it
is a follow-up commit so the refactor stays atomic (HIGH-9 step 1 of 2).
Closes HIGH-9 (step 1: extraction)
Refs HIGH-9 (step 2: manager.ts integration deferred until verification)
Co-authored-by: manager-extract (deep / gpt-5.3-codex high)
YeonGyu-Kim