Commit Graph

1488 Commits

Author SHA1 Message Date
YeonGyu-Kim 1e7a7600a2 docs(agents-md): regenerate hierarchical AGENTS.md knowledge base for v4.1.1
Refresh all AGENTS.md files to reflect codebase state at 5ffbe0e24 (was cd31d2a1a, 197 commits behind).

Key drift corrections across 45 modified + 1 new file:

Root AGENTS.md:
- TS file counts: 1967 -> 2034 in src/ (1337 source + 697 test)
- LOC: 278k -> 292k
- Barrel index.ts: 120 -> 122
- Hook tier composition: 52/59 -> 54/61 (base/with team-mode)
- Tool Guard hooks: 14 -> 16 (add fsync-skip-warning, bash-file-read-guard)
- Add boulder feature, agent-ordering schema, .agents/ directory, v4.1.1 release tag
- Add generated/ directory entry

src/AGENTS.md:
- Subsystem inventory: agents 96->102, hooks 570->581, tools 306->314,
  features 389->400, shared 258->278, cli 150->158, plugin 55->56
- LOC totals refreshed for every subsystem
- Schema files: 32 -> 30

src/hooks/AGENTS.md:
- Tier 2 (Tool Guard): 14 -> 16 hooks, add fsyncSkipWarning row
- Total: 52 base / 59 team-mode -> 54 base / 61 team-mode
- zauc-mocks count: 7 -> 5

src/features/AGENTS.md:
- background-agent: 47 -> 57 files, mention archive fallback
- opencode-skill-loader: 33 -> 30
- tmux-subagent: 34 -> 32

src/plugin/AGENTS.md:
- Tool Guard composer count: 14 -> 16
- Aggregator total: 43 -> 45

src/cli/AGENTS.md:
- Add new boulder subcommand (BoulderState inspector)
- Command count: 6 -> 7

NEW: src/features/boulder-state/AGENTS.md
- Document the new Boulder work tracking feature
- Schema v2 with BoulderState/BoulderWorkState/TaskSessionState
- Lifecycle, storage, integration points with atlas/ralph-loop hooks

All other AGENTS.md files: Generated date 2026-05-08 -> 2026-05-14.
2026-05-14 12:57:46 +09:00
PeterPonyu 5b99a87c6a fix(runtime-fallback): preserve attemptCount when our own abort is the cause (closes #4006)
When runtime-fallback aborts an in-flight request to swap in a fallback
model, opencode emits session.error{isAbort:true} as a consequence. The
existing event handler treated that as a user cancellation and called
resetRetryState — wiping attemptCount. Every subsequent provider
auto-retry signal then started over at attempt:1, never reaching
max_fallback_attempts, producing an infinite retry loop firing a new
fallback every ~2 seconds.

The bug only surfaces when the configured fallback target itself
silently fails (e.g. github-copilot quota exhausted): the original
model keeps re-emitting retry signals, our handler keeps "fixing"
them, the counter never advances. Reproducible on upstream/dev HEAD
(5ffbe0e24e).

Fix:
- New `internallyAbortedSessions: Set<string>` on HookDeps tracks
  sessions whose abort we triggered ourselves.
- abortSessionRequest in auto-retry.ts adds the session to the set
  when called with one of our internal sources:
  "session.status.retry-signal", "message.updated.retry-signal",
  "session.timeout". The "session.stop" source (user-initiated) is
  intentionally NOT marked — that path must still wipe state.
- handleSessionError in event-handler.ts checks the set before the
  cancellation branch. If the session is marked, consume the flag
  (delete it so a later user-abort still gets the reset) and skip
  resetRetryState. The state's attemptCount is preserved, so the
  next iteration progresses 1→2→3→... until max_fallback_attempts.
- dispose() clears the new set alongside the other per-session maps.

Tests: 3 new event-handler integration tests cover the fix
(internal-abort preserves state, external-abort still resets,
consecutive internal-abort cycles advance attemptCount). Existing
tests pass: 7/7 on event-handler. Pre-existing 2 dispose-test flakes
on the full runtime-fallback suite were verified to exist on
upstream/dev without this patch — unrelated.

bun run build: pass. bun run typecheck: pass.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 23:55:52 -04:00
YeonGyu-Kim 5ffbe0e24e fix(fallback): guard duplicate prompt injections 2026-05-14 01:03:19 +09:00
YeonGyu-Kim 47d60a74d3 fix(non-interactive-env): honor Windows ComSpec shell
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-13 22:47:37 +09:00
YeonGyu-Kim fef1d4530b fix(non-interactive-env): respect Windows command shell
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-13 20:07:46 +09:00
YeonGyu-Kim 61ba4e3b41 fix(ralph-loop): guard delayed start snapshots
Prevent delayed loop-start message counts from overwriting active Ralph Loop state after the loop has already advanced, so ULW completion can still enter Oracle verification instead of iterating forever.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-13 18:59:07 +09:00
YeonGyu-Kim 3b4d2431ee fix(hooks): guard stale idle prompts 2026-05-13 17:02:17 +09:00
YeonGyu-Kim a337635e3b fix(background-agent): defer active parent wakes 2026-05-13 17:02:17 +09:00
YeonGyu-Kim 1189b96d42 fix(continuation): mark atlas resumes synthetic 2026-05-13 13:25:01 +09:00
YeonGyu-Kim 38b1433ff5 fix(continuation): mark resumes synthetic 2026-05-13 13:09:26 +09:00
Ivan Smetanin 83c3379b5a Merge branch 'dev' into fix/surface-subagent-quota-error 2026-05-12 22:17:38 +01:00
tw-yshuang 12f5233877 merge(dev): resolve latest manager and runtime-fallback conflicts
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>
2026-05-12 22:49:57 +08:00
YeonGyu-Kim 75825eb9a6 fix(todo-description-override): add OpenCode schema contract for string priorities 2026-05-12 18:18:52 +09:00
YeonGyu-Kim d5fbada13d test: make unsafe test coercion explicit
Move test coercion out of a hidden global and require each test to import the helper so review tools and runtime scripts can see the unsafe boundary.

Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-12 15:51:31 +09:00
YeonGyu-Kim 0787867384 test(ralph-loop): remove unsafe fixture assertions
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-12 15:41:17 +09:00
YeonGyu-Kim 23211159c9 test(hooks): remove unsafe test assertions
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-12 15:41:17 +09:00
YeonGyu-Kim 9ffef79dbb fix(ralph-loop): ignore synthetic idle replays 2026-05-12 15:28:46 +09:00
YeonGyu-Kim 2c0f8e4645 fix(ralph-loop): return prompt dispatch failures 2026-05-12 13:14:07 +09:00
YeonGyu-Kim 2d656bd6bf Merge pull request #3964 from code-yeongyu/fix/electron-bun-runtime-shims
fix(electron): eliminate raw Bun.* runtime calls so plugin loads on OpenCode Desktop
2026-05-12 13:00:49 +09:00
YeonGyu-Kim 0aafe20a85 refactor: route raw Bun.file/write/hash/which/spawn through runtime shims
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.
2026-05-12 12:46:50 +09:00
YeonGyu-Kim 4da48555ee fix(plugin): normalize event session ids
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.
2026-05-12 12:32:26 +09:00
YeonGyu-Kim 0de6bd7253 fix(ralph-loop): reject promptAsync error responses 2026-05-12 12:06:49 +09:00
tw-yshuang 1e1f574b95 merge(dev): resolve latest sync-task conflict for delegated fallback PR
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>
2026-05-12 00:37:01 +08:00
Ivan Smetanin ecb92c1e70 fix(runtime-fallback): abort stuck subagent on quota error with no fallback
When a subagent session (e.g. Momus on GPT) hits a quota/usage-limit
error and the agent's category has no `fallback_models` configured, the
runtime-fallback hook previously returned silently. The OpenCode session
stayed in `retry` status indefinitely while the SDK kept hitting the
limit, the sync-task poller treated `retry` as active work, and the
parent's pending `task` tool call never resolved — leaving a stuck
"waiting for subagent" indicator in the parent conversation.

Narrow fix: at the `fallbackModels.length === 0` exit point, if the
session is a known subagent AND the error classifies as
`quota_exceeded`, abort the subagent session. The existing
`getTerminalSessionError` path in `sync-session-poller.ts` then surfaces
the error via the parent's tool result, which is the persistent surface
the user is already watching.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 14:40:43 +01:00
YeonGyu-Kim 71a63c20ec Revert "fix(hooks): dedupe native agent instructions"
This reverts commit 78d52872f2.
2026-05-11 18:02:13 +09:00
YeonGyu-Kim 78d52872f2 fix(hooks): dedupe native agent instructions 2026-05-11 17:21:58 +09:00
YeonGyu-Kim c849d0cbbc fix(ralph-loop): suppress stale iteration toasts 2026-05-11 15:17:07 +09:00
YeonGyu-Kim 2c299d2d9e Merge pull request #3943 from code-yeongyu/feature/boulder-evolution-and-discipline-agents
feat: boulder evolution + discipline agents (multi-work, timings, CLI, hooks, Oracle phase gates, no-excuses retry)
2026-05-11 14:54:58 +09:00
YeonGyu-Kim 29c42485a8 fix(hooks/atlas): capture plan snapshot for .sisyphus paths
Oracle review of PR #3943 surfaced that endTaskTimer never fires for real Prometheus plans because their canonical path is .sisyphus/plans/ and the snapshot capture was nested inside the !isSisyphusPath branch intended for direct-work warning suppression. Move the snapshot/path tracking out of the warning gate so all plan-file edits are snapshotted regardless of .sisyphus prefix. Keep the warning branch isSisyphus-gated so Atlas does not yell at legitimate plan edits.

Regression test now uses a real .sisyphus/plans/ path and fails against HEAD before the fix.
2026-05-11 14:48:49 +09:00
YeonGyu-Kim cf5fe757df feat(hooks/atlas): parse task_key from delegation prompt for parallel batches 2026-05-11 14:28:56 +09:00
YeonGyu-Kim e3cddb3650 feat(hooks/atlas): end task timer when plan checkbox flips to checked via edit 2026-05-11 14:27:03 +09:00
YeonGyu-Kim b8c25b3b75 refactor(hooks/atlas): remove unused resolveSessionOrigin helper 2026-05-11 14:26:10 +09:00
YeonGyu-Kim 079a2cd65a fix(start-work): preserve existing works when starting an explicit new plan 2026-05-11 14:25:52 +09:00
YeonGyu-Kim de9c28a095 fix(hooks/atlas): align completion behavior tests with task-4 timing updates 2026-05-11 13:49:36 +09:00
YeonGyu-Kim 1ebf89cb9f feat(hooks/atlas): inject boulder-complete elapsed-time nudge once per work
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-11 13:43:40 +09:00
YeonGyu-Kim a1c6e6b77d fixup! feat(hooks/atlas): use getWorkForSession in boulder lookups and session tracking 2026-05-11 13:42:39 +09:00
YeonGyu-Kim d6f4199cab feat(start-work): use getWorkResumeOptions for multi-work resume selection
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-11 13:40:57 +09:00
YeonGyu-Kim 29b44fffd0 feat(hooks/atlas): call completeBoulder when progress.isComplete 2026-05-11 13:40:48 +09:00
YeonGyu-Kim 127112e1e2 feat(hooks/atlas): wire per-task timers via startTaskTimer/endTaskTimer 2026-05-11 13:39:47 +09:00
YeonGyu-Kim f2a5ef0966 feat(hooks/atlas): add BOULDER_COMPLETE_PROMPT template and SessionState guard
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-05-11 13:37:35 +09:00
YeonGyu-Kim 18af3d3617 feat(hooks/atlas): use getWorkForSession in boulder lookups and session tracking 2026-05-11 13:37:19 +09:00
YeonGyu-Kim 49ff4b5f8d fix(compaction): ignore compaction agent updates
Fixes #3819
2026-05-11 13:25:29 +09:00
YeonGyu-Kim 2e36a92c25 fix(ralph-loop): revalidate ownership and harden commit/session-creation failures
Three correctness fixes on top of the dispatch-before-commit invariant:

- ralph-loop-event-handler.ts: after idleSettleMs, also require state ownership and non-verification-pending to match the event source before dispatching. Applied to both the session.idle and session.error retry paths.
- verification-failure-handler.ts: if incrementIteration fails after a successful continuation injection, clear the loop state and emit a warning toast instead of returning success.
- session-reset-strategy.ts: catch thrown session.create errors so they route through the typed session_creation_rejected path instead of surfacing as an unhandled rejection.
2026-05-11 13:02:16 +09:00
YeonGyu-Kim 35cab4db10 test(ralph-loop): cover ownership race, verification commit failure, session.create throw
Three additional invariant tests addressing the gaps surfaced by Cubic and the post-implementation review:

- idle path must not dispatch when state ownership changes during the idleSettleMs window
- verification-failure path must treat incrementIteration failure as a loud failure, not a success
- reset strategy must surface session.create rejections as session_creation_rejected even when the SDK throws instead of returning an error envelope
2026-05-11 13:02:07 +09:00
YeonGyu-Kim caaae19155 fix(ralph-loop): commit iteration only after verification continuation dispatches
Split the verification-failure restart into clearVerificationState (clears\nthe verification flags so we cleanly transition back to the main loop)\nfollowed by injectContinuationPrompt, with incrementIteration only on\nsuccessful injection. On rejection: clear the loop state and emit a loud\nwarning toast. Mirrors the dispatch-before-commit contract enforced for\nthe idle and session.error paths.
2026-05-11 12:46:35 +09:00
YeonGyu-Kim 09c45c3acb fix(ralph-loop): commit iteration only after continuation is dispatched
Reorder the session.idle and session.error retry paths so the durable\niteration counter and the progress toast advance only when continueIteration\nreturns dispatched. On dispatch_rejected or session_creation_rejected,\nclear the loop state and emit a loud failure toast instead of silently\nlogging while the loop appears to make progress.\n\nAdds an explicit settle-window state check so a session.deleted firing\nduring the idleSettleMs sleep no longer feeds dispatch against a cleared\nloop. Keeps idleSettleMs intact for the original idle-settle race.
2026-05-11 12:46:07 +09:00
YeonGyu-Kim d4cdeaccbe fix(ralph-loop): return typed ContinuationResult from continueIteration
Replace silent returns in continueIteration with a discriminated union\n(dispatched | session_creation_rejected | dispatch_rejected). Wraps\ninjectContinuationPrompt in try/catch so reset-strategy createIterationSession\nreturning null and promptAsync rejections both surface as typed failures\nthe caller can react to.
2026-05-11 12:45:59 +09:00
YeonGyu-Kim f0857a88fb test(ralph-loop): add dispatch-failure invariant tests
Lock the contract that durable iteration state and visible UI must only\nadvance when the continuation dispatch is semantically accepted. Adds 4\npermanent invariant tests covering the idle, session.error retry, and\nverification-failure orchestration paths, plus the reset-strategy\nsilent-null path.
2026-05-11 12:45:51 +09:00
Qihao 583fa2420f fix(runtime-fallback): classify localized balance failures 2026-05-11 11:06:06 +08:00
tw-yshuang 74171b9156 merge(dev): resolve background-agent delegated fallback conflicts
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>
2026-05-11 03:02:14 +08:00