4a029258a4
fix: resolve 5 remaining pre-publish blockers (14, 15, 17, 21, 25c)
...
- completion-promise-detector: restrict to assistant text parts only,
remove tool_result from completion detection (blocker 14)
- ralph-loop tests: flip tool_result completion expectations to negative
coverage, add false-positive rejection tests (blocker 15)
- skill tools: merge nativeSkills into initial cachedDescription
synchronously before any execute() call (blocker 17)
- skill tools test: add assertion for initial description including
native skills before execute() (blocker 25c)
- docs: sync all 4 fallback-chain docs with model-requirements.ts
runtime source of truth (blocker 21)
Verified: bun test (4599 pass / 0 fail), tsc --noEmit clean
2026-03-28 15:57:27 +09:00
d2c576c510
fix: resolve 25 pre-publish blockers
...
- postinstall.mjs: fix alias package detection
- migrate-legacy-plugin-entry: dedupe + regression tests
- task_system: default consistency across runtime paths
- task() contract: consistent tool behavior
- runtime model selection, tool cap, stale-task cancellation
- recovery sanitization, context-limit gating
- Ralph semantic DONE hardening, Atlas fallback persistence
- native-skill description/content, skill path traversal guard
- publish workflow: platform awaited via reusable workflow job
- release: version edits reapplied before commit/tag
- JSONC plugin migration: top-level plugin key safety
- cold-cache: user fallback models skip disconnected providers
- docs/version/release framing updates
Verified: bun test (4599 pass), tsc --noEmit clean, bun run build clean
2026-03-28 15:24:18 +09:00
95801a4850
fix(ralph-loop): extract text from parsed entry instead of testing raw JSONL
2026-03-26 18:16:07 +09:00
774d0bd84d
fix(ralph-loop): restrict semantic completion to DONE promise and assistant entries
...
- Gate semantic detection on promise === 'DONE' in both transcript and
session message paths to prevent false positives on VERIFIED promises
- Restrict transcript semantic fallback to assistant/text entries only,
skipping tool_use/tool_result to avoid matching file content
- Add regression test for VERIFIED promise not triggering semantic detection
2026-03-25 16:49:53 +09:00
aaaeb6997c
fix(ralph-loop): add semantic completion detection as fallback for natural language ( fixes #2489 )
2026-03-23 20:46:10 +09:00
f6c24e42af
fix(ralph-loop): detect promise tags in tool_result parts for ulw verification
...
Oracle's <promise>VERIFIED</promise> arrives as a tool_result part from the
task() tool call, not as a text part. Both detectCompletionInSessionMessages
and collectAssistantText only scanned type=text parts, missing the
verification signal entirely. This caused ulw loops to fail verification
even when Oracle successfully emitted VERIFIED.
Include tool_result parts in promise detection alongside text parts.
Exclude tool_use parts to avoid false positives from instructional text.
2026-03-18 19:03:30 +09:00
ce8957e1e1
fix(ralph-loop): harden oracle verification flow
...
Capture oracle verification sessions more reliably and accept parent-session VERIFIED evidence so ULW loops do not retry after successful review.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-18 17:45:59 +09:00
3caae14192
fix(ralph-loop): abort stale Oracle sessions before ulw verification restart
...
When Oracle verification fails in ulw-loop mode, the previous Oracle
session was never aborted before restarting. Each retry created a new
descendant session, causing unbounded session accumulation and 500
errors from server overload.
Now abort the old verification session before:
- restarting the loop after failed verification
- re-entering verification phase on subsequent DONE detection
2026-03-18 12:49:27 +09:00
4bc7b1d27c
fix(ulw-loop): add fallback for Oracle verification session tracking
...
The verification_session_id was never reliably set because the
prompt-based attempt_id matching in tool-execute-after depends on
metadata.prompt surviving the delegate-task execution chain. When
this fails silently, the loop never detects Oracle's VERIFIED
emission.
Add a fallback: when exact attempt_id matching fails but oracle
agent + verification_pending state match, still set the session ID.
Add diagnostic logging to trace verification flow failures.
Add integration test covering the full verification chain.
2026-03-17 16:21:40 +09:00
1812c9f054
test(ralph-loop): cover overlapping ultrawork loops
...
Lock down stale-session and overwrite cases so a previous ULW verification flow cannot complete or mutate a newer loop.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-12 17:05:02 +09:00
f31537f14c
fix(ralph-loop): continue ultrawork until oracle verifies
...
Keep /ulw-loop iterating after the main session emits DONE so completion still depends on an actual Oracle VERIFIED result.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-12 17:00:25 +09:00
f4f54c2b7f
test(ralph-loop): remove volatile tool result timestamp
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-12 01:07:50 +09:00
8ec2c44615
fix(ulw-loop): retry parent session after failed verification
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-07 05:46:05 +09:00
898b628d3d
fix(ulw-loop): track Oracle verification sessions explicitly
...
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-03-06 22:37:41 +09:00
c17f7215f2
test(ulw-loop): cover Oracle verification flow
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 22:00:21 +09:00
a010de1db2
feat(ulw-loop): require Oracle verification before completion
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 22:00:14 +09:00
23dcd99c9a
docs(agents): refresh generated AGENTS guides
...
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-03-06 17:59:05 +09:00
33d39597ae
docs(agents): regenerate AGENTS.md hierarchy with updated metrics and model configs
...
- 1208→1243 TS files (+35), 143k→155k LOC (+12k)
- Update all agent models: Sisyphus, Hephaestus, Oracle, Librarian, Atlas, Metis, Momus
- Add 6 new hook directories (39→45 dirs): beast-mode-system, hashline-edit-diff-enhancer, anthropic-image-context, task-reminder, compaction-todo-preserver, runtime-fallback
- Update category models: visual-engineering/artistry gemini-3-pro→gemini-3.1-pro
- Add 2 config schema files: fallback-models.ts, runtime-fallback.ts
- Timestamp: 2026-03-02 | Commit: 1c2caa09🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-03-02 23:40:38 +09:00
f2a1412bf1
test(ralph-loop): harden completion detector PluginInput mock
2026-02-27 03:12:22 +09:00
190c6991ac
fix(ralph-loop): persist session message count at loop start
2026-02-27 03:08:30 +09:00
e17a00a906
fix(ralph-loop): scope completion detection to messages since loop start
2026-02-27 03:05:14 +09:00
2295161022
fix(ralph-loop): add inFlight guard and improve completion detection to prevent infinite loops
...
Closes #2084
2026-02-26 20:58:55 +09:00
a0e57c13c3
fix(ralph-loop): prevent race condition in reset strategy between session ID update and TUI switch
...
Fixes #2100
2026-02-25 14:01:27 +09:00
73453a7191
docs(agents): update hook counts 44→46, add hashline-edit documentation
...
- Update root AGENTS.md: hook count 44→46, commit fcb90d92🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-02-25 00:02:05 +09:00
fe415319e5
fix: resolve publish blockers for v3.7.4→v3.8.0 release
...
- Fix #1991 crash: optional chaining for task-history sessionID access
- Fix #1992 think-mode: add antigravity entries to HIGH_VARIANT_MAP
- Fix #1949 Copilot premium misattribution: use createInternalAgentTextPart
- Fix #1982 load_skills: pass directory to discoverSkills for project-level skills
- Fix command priority: sort scopePriority before .find(), project-first return
- Fix Google provider transform: apply in userFallbackModels path
- Fix ralph-loop TUI: optional chaining for event handler
- Fix runtime-fallback: unify dual fallback engines, remove HTTP 400 from retry,
fix pendingFallbackModel stuck state, add priority gate to skip model-fallback
when runtime-fallback is active
- Fix Prometheus task system: exempt from todowrite/todoread deny
- Fix background_output: default full_session to true
- Remove orphan hooks: hashline-edit-diff-enhancer (redundant with hashline_edit
built-in diff), task-reminder (dead code)
- Remove orphan config entries: 3 stale hook names from Zod schema
- Fix disabled_hooks schema: accept arbitrary strings for forward compatibility
- Register json-error-recovery hook in tool-guard pipeline
- Add disabled_hooks gating for question-label-truncator, task-resume-info,
claude-code-hooks
- Update test expectations to match new behavior
2026-02-21 16:24:18 +09:00
940e49b44c
fix(ralph-loop): use shared isRecord, fix quoted argument parsing for prompt and completion-promise
2026-02-21 05:36:11 +09:00
590dc04be7
fix(ralph-loop): bind selectSession to tui context, use sourceSessionID for tool inheritance, handle flag-only arguments, fix test provider mocks
2026-02-21 05:36:11 +09:00
7bb427078a
fix(ralph-loop): use inherited fallback context and SDK TUI session selection
2026-02-21 05:33:53 +09:00
5c83fee619
feat(ralph-loop): add strategy option for fresh context per iteration
...
Closes #1901
Add 'default_strategy' config option (default: 'continue') to control whether ralph-loop creates a new session per iteration ('reset') or keeps the same session ('continue'). The 'reset' strategy keeps the model in the smart zone by starting with fresh context for each iteration.
Supports --strategy flag for per-command override.
2026-02-21 05:33:53 +09:00
b1eccf7425
docs: update all 31 AGENTS.md files with current project state
2026-02-21 04:38:18 +09:00
850fb0378e
fix(copilot): mark internal hook injections as agent-initiated
...
Apply the internal initiator marker to automated continuation, recovery, babysitter, stop-hook, and hook-message injections so Copilot attribution consistently sets x-initiator=agent for system-generated prompts.
2026-02-19 13:17:02 +02:00
2034cf137a
docs: add module-level AGENTS.md for config-manager, keyword-detector, ralph-loop, session-recovery, todo-continuation-enforcer
...
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-02-19 10:46:40 +09:00
096db59399
fix(run): inherit main-session tool permissions for continuation prompts
2026-02-18 21:10:21 +09:00
1a6810535c
refactor: create normalizeSDKResponse helper and replace scattered patterns across 37 files
2026-02-16 18:20:19 +09:00
c9c02e0525
refactor(shared): consolidate 13+ getMessageDir copies into single shared function
2026-02-16 16:13:39 +09:00
88e1e3d0fa
fix(ralph-loop): only scan text parts for completion tags and handle both API shapes
...
Reasoning parts could contain completion-like text triggering false
positives. Also handles session.messages returning either an array
or {data: [...]} shape.
2026-02-11 00:45:51 +09:00
f3f6ba47fe
merge: integrate origin/dev into modular-enforcement branch
...
Resolves all merge conflicts, preserving our split module structure
while integrating all dev changes:
- Custom agent summaries support (parseRegisteredAgentSummaries)
- Background notification queue (enqueueNotificationForParent)
- Atlas shared git-worktree module (collectGitDiffStats, formatFileChanges)
- Ralph-loop withTimeout + DEFAULT_API_TIMEOUT=5000
- Session recovery assistant_prefill_unsupported error type
- Atlas agentOverrides forwarding
- Config handler plan model demotion (buildPlanDemoteConfig)
- Delegate-task agentOverrides, promptSyncWithModelSuggestionRetry, variant
- LSP init timeout + stale init detection
- isPlanFamily function + task-continuation-enforcer hook
- Handoff command
2026-02-08 17:34:47 +09:00
0f145b2e40
refactor(ralph-loop): split hook into state controller and event handler modules
...
Extract Ralph loop lifecycle management:
- loop-state-controller.ts: start/stop/recovery state machine
- ralph-loop-event-handler.ts: event handling logic
- continuation-prompt-builder.ts, continuation-prompt-injector.ts
- completion-promise-detector.ts, loop-session-recovery.ts
- message-storage-directory.ts
2026-02-08 16:22:10 +09:00
119e18c810
refactor: wave 2 - split atlas, auto-update-checker, session-recovery, todo-enforcer, background-task hooks
...
- Extract atlas/ into 15 focused modules (hook, event handler, tool policies, types, etc.)
- Split auto-update-checker into checker/ and hook/ subdirectories with single-purpose files
- Decompose session-recovery into separate recovery strategy files per error type
- Extract todo-continuation-enforcer from monolith to directory with dedicated modules
- Split background-task/tools.ts into individual tool creator files
- Extract command-executor, tmux-utils into focused sub-modules
- Split config/schema.ts into domain-specific schema files
- Decompose cli/config-manager.ts into focused modules
- Rollback skill-mcp-manager, model-availability, index.ts splits that broke tests
- Fix all import path depths for moved files (../../ -> ../../../)
- Add explicit type annotations to resolve TS7006 implicit any errors
Typecheck: 0 errors
Tests: 2359 pass, 5 fail (all pre-existing)
2026-02-08 15:01:42 +09:00
676ff513fa
fix: detect completion tags in ralph/ULW loop to stop iteration ( #1233 )
2026-02-08 14:50:36 +09:00
b2661be833
test: fix ralph-loop tests by adding promptAsync to mock
...
The ralph-loop hook calls promptAsync in the implementation, but the
test mock only defined prompt(). Added promptAsync with identical
behavior to make tests pass.
- All 38 ralph-loop tests now pass
- Total test suite: 2361 pass, 3 fail (unrelated to this change)
2026-02-08 02:41:29 +09:00
46e02b9457
fix(hooks): switch session.prompt to promptAsync in all hooks
2026-02-07 13:42:24 +01:00
f146aeff0f
refactor: major codebase cleanup - BDD comments, file splitting, bug fixes ( #1350 )
...
* style(tests): normalize BDD comments from '// #given' to '// given'
- Replace 4,668 Python-style BDD comments across 107 test files
- Patterns changed: // #given -> // given, // #when -> // when, // #then -> // then
- Also handles no-space variants: //#given -> // given
* fix(rules-injector): prefer output.metadata.filePath over output.title
- Extract file path resolution to dedicated output-path.ts module
- Prefer metadata.filePath which contains actual file path
- Fall back to output.title only when metadata unavailable
- Fixes issue where rules weren't injected when tool output title was a label
* feat(slashcommand): add optional user_message parameter
- Add user_message optional parameter for command arguments
- Model can now call: command='publish' user_message='patch'
- Improves error messages with clearer format guidance
- Helps LLMs understand correct parameter usage
* feat(hooks): restore compaction-context-injector hook
- Restore hook deleted in cbbc7bd0justsisyphus@users.noreply.github.com >
2026-02-01 16:47:50 +09:00
6f348a8a5c
fix: resolve CI test timeouts with configurable timing
...
- Add timing.ts module for test-only timing configuration
- Replace hardcoded wait times with getTimingConfig()
- Enable all previously skipped tests (ralph-loop, session-state, delegate-task)
- Tests now complete in ~2s instead of timing out
2026-01-28 14:17:56 +09:00
20cca35157
fix(ralph-loop): skip user messages in transcript completion detection ( #622 ) ( #1086 )
...
* fix(ralph-loop): skip user messages in transcript completion detection (#622 )
The transcript-based completion detection was searching the entire JSONL
file for <promise>DONE</promise>, including user message entries. The
RALPH_LOOP_TEMPLATE instructional text contains this literal pattern,
which gets recorded as a user message, causing false positive completion
detection on every iteration. This made the loop always terminate at
iteration 1.
Fix: Parse JSONL entries line-by-line and skip entries with type 'user'
so only tool_result/assistant entries are checked for the completion
promise. Also remove the hardcoded <promise>DONE</promise> from the
template exit conditions as defense-in-depth.
* chore: changes by sisyphus-dev-ai
---------
Co-authored-by: sisyphus-dev-ai <sisyphus-dev-ai@users.noreply.github.com >
2026-01-25 12:34:42 +09:00
d13e8411f0
Add /ulw-loop command for ultrawork mode loop ( #867 )
...
* feat(ralph-loop): add ultrawork field to RalphLoopState
* feat(ralph-loop): persist ultrawork field in storage
* feat(ralph-loop): accept ultrawork option in startLoop
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
* feat(ralph-loop): prepend ultrawork keyword when mode active
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
* feat(ralph-loop): custom toast for ultrawork mode
* feat(ralph-loop): add /ulw-loop command for ultrawork mode
* fix(ralph-loop): add non-null assertion for type safety
* fix(ralph-loop): mirror argument parsing in ulw-loop handler
- Parse quoted prompts and strip flags from task text
- Support --max-iterations and --completion-promise options
- Add default prompt for empty input
- Fixes behavior inconsistency with /ralph-loop
---------
Co-authored-by: justsisyphus <sisyphus-dev-ai@users.noreply.github.com >
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-01-17 19:56:50 +09:00
d00c2e7439
fix(hooks): extract model from assistant messages with flat modelID/providerID
...
OpenCode API returns different structures for user vs assistant messages:
- User: info.model = { providerID, modelID } (nested)
- Assistant: info.modelID, info.providerID (flat top-level)
Previous code only checked nested format, causing model info loss when
continuation hooks fired after assistant messages.
Files modified:
- todo-continuation-enforcer.ts
- ralph-loop/index.ts
- sisyphus-task/tools.ts
- background-agent/manager.ts
Added test for assistant message model extraction.
2026-01-16 13:54:22 +09:00
c7ca608b38
refactor: unify system directive prefix for keyword-detector filtering
...
- Add shared/system-directive.ts with SYSTEM_DIRECTIVE_PREFIX constant
- Unify all system message prefixes to [SYSTEM DIRECTIVE: OH-MY-OPENCODE - ...]
- Add isSystemDirective() filter to keyword-detector to skip system messages
- Update prometheus-md-only tests to use new prefix constants
2026-01-16 11:35:37 +09:00
e264cd5078
test: skip flaky timeout test in CI
2026-01-16 02:45:41 +09:00
c9f762f980
fix(hooks): use API instead of filesystem to resolve model info for session.prompt
...
Previously, continuation hooks (todo-continuation, boulder-continuation, ralph-loop)
and background tasks resolved model info from filesystem cache, which could be stale
or missing. This caused session.prompt to fallback to default model (Sonnet) instead
of using the originally configured model (e.g., Opus).
Now all session.prompt calls first try API (session.messages) to get current model
info, with filesystem as fallback if API fails.
Affected files:
- todo-continuation-enforcer.ts
- sisyphus-orchestrator/index.ts
- ralph-loop/index.ts
- background-agent/manager.ts
- sisyphus-task/tools.ts
- hook-message-injector/index.ts (export ToolPermission type)
2026-01-16 02:33:44 +09:00
YeonGyu-Kim