When many background tasks complete in rapid succession while the parent
session is idle, each completion fired its own promptAsync call, stacking
N consecutive `<system-reminder>` user messages with no assistant turn
between. Hyperplan + many parallel explore subagents made this very
visible to the user.
Route the idle-path through the existing pendingParentWakes queue with a
100ms debounce window. Notifications arriving during the debounce join
the same batch, the 150ms settle window also coalesces newcomers, and a
single batched prompt fires to the parent. Busy-path semantics are
unchanged (still 1s retry).
Prior attempts (1c05c60dc, ea55c385b, a337635e3) all coalesced only the
busy-defer path, leaving the idle-immediate-send path uncoalesced.
Reframe the skill's Lifecycle section so the lead treats teams as
ephemeral, one-per-phase units. The moment a phase ends or the shape
no longer fits, call team_delete and spawn a fresh team. Restructure
through delete-then-create, never in place.
Also fixes a misframing in old step 5: team_shutdown_request is a
per-session self-shutdown signal, not a 'wind down the team' command.
team_delete is what tears the whole team down.
Sync the PR branch with the newest dev branch and resolve the new import-level conflicts in background-agent manager and runtime-fallback tests. Preserve both the delegated bootstrap coverage from this branch and the newer upstream test utilities and runtime wiring changes, then re-verify the affected delegated fallback suites and typecheck.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
The circuit breaker in manager.ts uses recordToolCall to detect when a subagent gets stuck repeating identical tool_use blocks. It passed partInfo.state?.input as the tool-input signature. When a model (Kimi K2.6 in the reporter's case) emits duplicate tool_use parts faster than the tool actually starts running, state.input is still null, so loop-detector falls back to the bare 'tool::__unknown-input__' signature. As soon as one part has state.input populated (next event), the signature flips to 'tool::{actual-args}' and the consecutive counter resets to 1, repeatedly. The breaker never reaches its 20-call threshold.
Add a top-level input?: Record<string, unknown> field to the local MessagePartInfo interface and prefer state.input when present, falling back to the part's own input when state is still pre-running. The OpenCode part payload carries the tool input as soon as the tool_use block is generated, so this fallback restores signature stability across the model's repeated emissions.
Verification: added 2 regression tests in manager-circuit-breaker.test.ts. Test 1 (reproduce) emits 20 part.updated events with only top-level input and asserts the task is cancelled by the breaker — fails before the fix, passes after. Test 2 confirms that when state.input IS present, it still wins over the top-level input (precedence preserved). All 10 manager-circuit-breaker tests pass, all 20 loop-detector tests pass, typecheck clean.
Cubic AI reviewer flagged the use of the untrusted Host header as the
URL base in startCallbackServer. The server only binds to 127.0.0.1,
so hardcoding "http://127.0.0.1" as the URL base is robust against
malformed or manipulated Host values and matches upstream behavior
prior to the node:http refactor.
The OAuth callback server was using Bun.serve, which would crash if
mcp-oauth code paths ever entered the plugin bundle. Switch to
node:http.createServer with the same WHATWG behavior:
- Binds to 127.0.0.1, preserves 200/400/404 status codes
- Translates fetch(Request)→Response to (req, res) callback style
- Clears OAuth timeout on success/error/missing-param paths
- Replaces server.stop(true) with server.close() for shutdown
Functional behavior and response bodies unchanged.
Eliminates 19 unguarded `Bun.*` runtime call sites in the plugin bundle
that crashed with `ReferenceError: Bun is not defined` under Electron.
Per-tool-call hot paths (executed on every Read/Edit):
- src/tools/hashline-edit/hash-computation.ts: Bun.hash.xxHash32 → bunHashXxh32
- src/tools/hashline-edit/hashline-edit-executor.ts: 8 sites via bunFile/bunWrite
- src/hooks/hashline-read-enhancer/hook.ts: Bun.file → bunFile
- src/hooks/hashline-edit-diff-enhancer/hook.ts: 2 sites via bunFile
Plugin-load paths:
- src/hooks/claude-code-hooks/config.ts and config-loader.ts: Bun.file → bunFile
- src/features/claude-code-mcp-loader/loader.ts: Bun.file → bunFile
- src/features/claude-code-plugin-loader/mcp-server-loader.ts: Bun.file → bunFile
- src/features/team-mode/deps.ts: Bun.spawn → spawn shim
- src/hooks/session-notification-utils.ts: Bun.which → bunWhich, also drops
the bare `declare const Bun` ambient declaration
- src/shared/binary-downloader.ts: Bun.write → bunWrite
Pure mechanical API swaps. No control-flow or signature changes.
Handle OpenCode session events that carry the session ID under properties.info.id or properties.info.sessionID so background tasks and continuation hooks do not miss idle/error/delete events.
Add regression coverage for nested session.idle events completing background tasks and waking continuation hooks.
Sync the PR branch with the latest dev branch and resolve the remaining conflict in sync-task.test.ts while preserving both the new upstream poll-recovery coverage and this branch's delegated bootstrap cleanup and isolation coverage. Re-verified the affected delegated fallback suites and typecheck after the merge resolution.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Currently `registerManagerForCleanup` unconditionally installs global `uncaughtException` and `unhandledRejection` listeners that call `process.exit(1)` after cleanup. For users who load the plugin but never run background-agent tasks, these handlers turn transient streaming errors (e.g. undici `UND_ERR_SOCKET` mid-stream resets from `api.githubcopilot.com`) into a full process kill — opencode dies after every flaky response.
Add an `OMO_DISABLE_PROCESS_CLEANUP` env var (accepts 1/true/yes/on, case-insensitive) that skips just the error-event registration. Signal handlers (SIGINT/SIGTERM/SIGBREAK/beforeExit/exit) remain installed so graceful shutdown of any in-flight cleanup targets still runs. This is the lowest-risk near-term mitigation suggested in the issue (option #2): users opting in pay the cost of unhandled rejections themselves, but no longer lose their session to a transient socket reset.
Verification: 6 new test cases cover env-var precedence (set/unset, truthy/falsy values), signal-handler preservation, and behavior under `uncaughtException`. All 20 tests in process-cleanup.test.ts pass. Typecheck clean. Manual QA confirms env-var detection works end-to-end.
The agent-not-found fallback tests use fire-and-forget prompt chains
that must settle before assertions run. A fixed 50ms setTimeout is
insufficient on slower CI runners, causing intermittent failures.
Replace with waitForCondition() that polls at 10ms intervals with a
2s timeout — fast on green runs, reliable on slow CI.
Closes#3935
Reconcile the latest dev branch changes with the delegated child-session fallback work. Preserve the upstream background-agent updates while keeping the delegated bootstrap cleanup and compatibility wiring fixes intact, then re-verify the affected regression suites and typecheck.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
MessageAbortedError/worker shutdown could race with scheduled removeTask, leaving background_output's manager.getTask returning 'Task not found' even though the task had completed cleanly.
Fix: add completedTaskArchive (max 500, FIFO eviction). On removeTask, archive non-running/pending tasks with sessionId. getTask falls back to archive on active-map miss. addTask clears stale archive entries on re-registration.
Fixes#3895
Sub-agent sessions created via the task tool or call_omo_agent tool were
being created without a model. The resolved model was only passed as a
promptAsync body override, which opencode core ignores, causing fallback
to the client's system default model (often a reasoning model that hangs).
OpenCode's session.create API supports model: { id, providerID, variant }
at creation time. This fix ensures the resolved category/agent model is
passed during session creation across all paths:
- createSyncSession() (delegate task / category-based)
- createOrGetSession() (call_omo_agent / direct agent calls)
- BackgroundManager.startTask() (background tasks)
This guarantees each sub-agent session is created with the correct model
regardless of whether promptAsync honors its model override.
Replaces direct fileHandle.sync() calls in acquireLock and atomicWrite
with tolerantFsync. Users on iCloud Drive / OneDrive / Desktop sync
folders were hitting 'EPERM: operation not permitted, fsync' during
team_create, which propagated up and aborted the entire team_create
flow even though the actual write+rename had succeeded.
Reported on Discord (omo 4.0.0, opencode desktop 1.14.41, project on
synced Desktop). atomicity is preserved by the temp-file rename; only
the durability hint is now best-effort on filesystems that disallow
fsync.
The previous commit (2dfa6336f) used 'git add -A' which swept up files that
prior commits had intentionally deleted but were still present untracked in
the local workspace. This commit re-deletes them to match the upstream intent.
Files removed (each was previously deleted in the cited commit):
- drafts/gpt-5-5/{README,deep,hephaestus,oracle,sisyphus-junior,sisyphus}.md
(deleted in 40ded422c chore(drafts): remove stale gpt-5.5 prompt drafts)
- src/plugin-dispose.{ts,test.ts}
(deleted in e2f5c0d36 refactor(plugin): remove orphaned createPluginDispose)
- src/features/tmux-subagent/{cleanup,session-created-handler,session-deleted-handler}.ts
(deleted in 7a7926f22 chore(tmux-subagent): remove dead event-handler modules)
- src/tools/delegate-task/{model-string-parser,resolve-call-id,resolve-call-id.test}.ts
(model-string-parser deleted in db056346d; resolve-call-id was scratch)
- src/__debug-test.test.ts (debug scratch never intended for git)
Typecheck + model-requirements tests still pass. The legitimate metis +
AGENTS.md edits from the previous commit remain on dev.
Source code change:
- src/shared/model-requirements.ts: prepend claude-sonnet-4-6 to metis fallback
chain so Sonnet becomes the default. Opus 4.7 max remains as the immediate
fallback for callers who want extra reasoning.
- src/shared/model-requirements.test.ts: update assertion to expect Sonnet
primary + Opus secondary.
AGENTS.md accuracy fixes (verified against source):
- Agent modes: Sisyphus/Hephaestus are 'primary' (not 'all'); Sisyphus-Junior
is 'subagent' (not 'all'). Confirmed via 'const MODE: AgentMode = ...' in
each agent file. Also clarified Prometheus has no agentSources factory and
is built via buildPrometheusAgentConfig.
- Sisyphus fallback chain: corrected order to kimi-k2.6 → k2p5 → kimi-k2.5
→ gpt-5.5 medium → glm-5 → big-pickle (was missing kimi-k2.5).
- Librarian/Explore: added missing minimax-m2.7 step between -highspeed and
claude-haiku-4-5.
- Metis chain: removed fictitious gemini-3.1-pro entry.
- Sisyphus-Junior chain: spelled out the actual fallback (was 'user-configurable').
- Temperatures: Sisyphus/Hephaestus do not set explicit temperature (model
default); Sisyphus-Junior is 0.1 via SISYPHUS_JUNIOR_DEFAULTS.
- Quick category default: gpt-5.4-mini (not gpt-5.4-mini-fast).
Team-mode corrections:
- Eligibility registry has 3 verdicts: eligible (sisyphus, atlas, sisyphus-junior),
conditional (hephaestus — needs D-36 teammate permission), hard-reject
(oracle, librarian, explore, multimodal-looker, metis, momus, prometheus).
- Schema has 11 fields, not 4: added max_messages_per_run, max_wall_clock_minutes,
max_member_turns, base_dir, message_payload_max_bytes, recipient_unread_max_bytes,
mailbox_poll_interval_ms.
- Hooks: 'team-session-events' is 4 sub-handlers in src/plugin/event.ts
(team-idle-wake-hint, team-lead-orphan-handler, team-member-error-handler,
team-member-status-handler), not a single Continuation-tier hook.
- Tier counts now show base + team-mode: ToolGuard 14/15, Transform 5/7.
- Total: 52 base hooks, 59 with team-mode.
Doc cascade for the Metis change:
- docs/guide/orchestration.md, agent-model-matching.md, installation.md
- docs/reference/configuration.md, features.md
Capture delegated child-session retry context before the first prompt so fallback recovery still works when session history is empty. Align background and sync launch paths around the same bootstrap contract, clear session-scoped fallback state on every terminal path, and lock the behavior with regression coverage for first-prompt retries, exhaustion, isolation, and cleanup.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
YeonGyu-Kim