a1ca658d76
fix(tool-config): stop overriding external_directory permission
...
Remove the hardcoded external_directory: "allow" default from
applyToolConfig(). This was silently overriding OpenCode's built-in
default of "ask" and any user-configured external_directory permission.
With this change, external_directory permission is fully controlled by
OpenCode's defaults and user configuration, as intended.
Fixes #1973
Fixes #2194
2026-03-06 17:58:08 +08:00
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
7718969317
feat(model-requirements): prefer GPT-5.4 and glm-5 in agent fallback chains
...
Align Prometheus, Momus, and Atlas with newer GPT-5.4 fallback tiers and replace Sisyphus install-time GLM-4.7 fallbacks with GLM-5 only.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 17:43:48 +09:00
7fe44024c0
feat(no-sisyphus-gpt): allow Sisyphus with GPT-5.4 model
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 17:35:31 +09:00
901ddda09c
refactor(sisyphus): extract prompt builders into subdirectory with GPT-5.4 variant
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 17:35:24 +09:00
b062fc45cb
fix: address Cubic P2 review - fake timers in tests, add opencode provider to glm-5
...
Replace real setTimeout(7000) with fake timer interception in atlas
retry tests (35s -> 227ms). Add missing opencode provider to glm-5
fallback in unspecified-high category.
2026-03-06 16:55:02 +09:00
4eb38d99d2
fix(atlas): add full eligibility checks to delayed retry callback
...
Address Cubic P1 review: timer callback now re-checks failure backoff
count, boulder session membership, and running background tasks before
injecting continuation, matching the main idle handler's eligibility
gate.
2026-03-06 16:31:48 +09:00
cecb78e944
fix(atlas): schedule delayed retry when cooldown blocks boulder continuation
...
When atlas injects a boulder continuation via promptAsync() and the
model's response is immediately aborted (MessageAbortedError), OpenCode
fires a burst of session.idle events within milliseconds. Atlas blocks
all of them due to the 5-second cooldown. After the burst, OpenCode
stops generating session.idle events (it's state-change based, not
periodic), leaving the session stuck forever.
Fix: When cooldown blocks an idle event for a boulder session with an
incomplete plan, schedule a one-shot setTimeout (cooldown + 1s) to
re-attempt injection. The timer callback re-checks boulder state, plan
progress, and continuation-stopped flag before injecting. Only one timer
per session is allowed (deduped via pendingRetryTimer field). Timers are
cleaned up on session.deleted and session.compacted events.
2026-03-06 16:14:24 +09:00
764ca0c51b
feat(hephaestus): add generic GPT prompt fallback with model-specific routing
...
Split monolithic hephaestus.ts into directory with model-specific prompt
variants (gpt-5-4.ts, gpt-5-3-codex.ts, gpt.ts) mirroring the
sisyphus-junior pattern. Generic gpt.ts uses pre-codex-tuning prompt as
fallback for non-specific GPT models.
Also adds isGpt5_4Model and isGpt5_3CodexModel helpers to types.ts.
2026-03-06 15:34:37 +09:00
f4eba51388
feat(sisyphus-junior): add model-specific GPT prompt routing (gpt-5-4, gpt-5-3-codex, generic gpt)
...
Split GPT prompt into three variants with model-based routing:
- gpt-5-4.ts: GPT-5.4 optimized (expert coding agent framing, prose-first)
- gpt-5-3-codex.ts: GPT-5.3-Codex optimized (Hephaestus-style Senior Engineer)
- gpt.ts: Generic GPT fallback (Hephaestus-style, for any other GPT model)
Routing: gpt-5.4 → gpt-5-4 | gpt-5.3-codex → gpt-5-3-codex | other GPT → gpt
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-03-06 14:50:01 +09:00
17f11a5fa6
feat(metis,momus): add QA scenario executability checks
...
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 14:26:14 +09:00
cde6566792
refactor(atlas): add Final Verification Wave to orchestration workflow
...
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 14:26:04 +09:00
2e4fd5843c
feat(model-requirements): update unspecified-high to gpt-5.4, add glm-5/k2p5/kimi-k2.5 fallbacks
...
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 14:25:47 +09:00
ae05e76ddf
fix(start-work): remove worktree setup prompt when unspecified, add strong worktree active instructions
...
When no worktree is specified in boulder, stop injecting 'Worktree Setup Required'
instructions. When worktree IS present, inject emphatic instructions ensuring the
agent and all subagents operate exclusively within the worktree directory.
2026-03-06 14:20:32 +09:00
381d7688ab
refactor(ultrawork): rename gpt5.2.ts to gpt.ts and align with 5.4 style
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:39 +09:00
3d0ccdd019
feat(momus): add GPT-5.4 variant prompt with model-based routing
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:32 +09:00
051737078e
feat(oracle): add GPT-5.4 variant prompt with model-based routing
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:26 +09:00
983b4d8ca7
refactor(prometheus): align GPT prompt with 5.4 system prompt style
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:20 +09:00
de8e5ea97f
refactor(atlas): align GPT prompt with 5.4 system prompt style
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:14 +09:00
285db926da
refactor(sisyphus-junior): align GPT prompt with 5.4 system prompt style
...
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 13:44:05 +09:00
ee3d88af9d
refactor(installer): remove dead Antigravity auth plugin code
...
The installer was writing Antigravity provider config and calling a no-op addAuthPlugins function. Since opencode-antigravity-auth is no longer auto-installed and OpenCode supports native Google/Gemini auth, all Antigravity-related installer code is dead. Gemini detection now checks for native google provider instead.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode )
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai >
2026-03-06 10:59:41 +09:00
89dc302403
update agent-browser skill to match upstream v0.16.3
...
Sync SKILL.md and inline template with vercel-labs/agent-browser v0.16.3.
Adds: native Rust daemon, diff commands, annotated screenshots, profiler,
keyboard type/inserttext, get styles, expanded locators (placeholder/alt/
title/testid/last), security options, config file support, iOS Simulator,
cloud providers (Browserbase/Browser Use/Kernel), session persistence,
CDP auto-connect, and state management commands.
2026-03-06 10:45:35 +09:00
389625cb20
Update constants.ts
2026-03-05 11:41:39 +01:00
e916d564a9
fix(background-task): clarify timeout unit is milliseconds in description
2026-03-05 09:05:29 +01:00
a61f8bb853
Update @opencode-ai/plugin and SDK to v1.2.x and align system transform handler signature
...
- Bump @opencode-ai/plugin ^1.1.19 → ^1.2.16, @opencode-ai/sdk ^1.1.19 → ^1.2.17
- Update system-transform handler input type to match new plugin contract (optional sessionID, required model)
- Add @opencode-ai/sdk override in bun.lock
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode )
2026-03-05 11:18:12 +09:00
c8c99445ea
fix(look-at): add catch block to prevent TUI crash on unexpected errors
2026-03-05 11:11:53 +09:00
39d94a4af6
fix(task): disambiguate background task_id metadata
2026-03-05 11:02:49 +09:00
acf4c46439
fix(task): align background output task_id with opencode contract
2026-03-05 11:02:49 +09:00
5cbf7828f0
fix(task): avoid pending sessionId metadata in background delegate output
2026-03-05 11:02:49 +09:00
d6b0e564bf
feat(delegate-task): unify TUI metadata by adding model field to all 5 executor paths
2026-03-04 18:31:19 +09:00
85690b69a8
test: update snapshots and assertions for kimi-k2.5-free removal
2026-03-04 11:33:11 +09:00
8c2dcb75cb
refactor: remove kimi-k2.5-free from all fallback chains and reorder multimodal-looker
...
kimi-k2.5-free is no longer available. Remove from all agent and category
fallback chains (sisyphus, multimodal-looker, prometheus, metis, atlas,
writing). Reorder multimodal-looker to: gpt-5.3-codex medium -> k2p5 ->
gemini-3-flash -> glm-4.6v -> gpt-5-nano.
2026-03-04 11:24:39 +09:00
1ef5c17c35
feat: make gpt-5.3-codex medium the primary model for multimodal-looker
...
GPT-5.3 Codex has strong multimodal capabilities. Promote it to first
candidate in multimodal-looker fallback chain, with gemini-3-flash
following (matching the ULW pattern of gpt-5.3-codex -> gemini).
2026-03-04 11:20:55 +09:00
f9da00d021
Merge pull request #2251 from code-yeongyu/fix/pr-1906-image-conversion
...
fix(look-at): temp dir cleanup, Windows compat, argument injection prevention
2026-03-03 00:49:25 +09:00
51a3d20dc9
Merge pull request #2250 from code-yeongyu/fix/pr-2113-model-fallback
...
fix(model-fallback): correct transform expectations and hermetic test isolation
2026-03-03 00:48:11 +09:00
785dd529e1
Merge pull request #2249 from code-yeongyu/fix/pr-2173-timeout-issues
...
fix(delegate-task): resolve timeout handling regressions from #2173
2026-03-03 00:48:08 +09:00
025d2a3579
Merge pull request #2248 from code-yeongyu/fix/pr-2080-model-format
...
fix: model format normalization and explicit config cache bypass
2026-03-03 00:48:04 +09:00
0e858ee1df
Merge pull request #2247 from code-yeongyu/fix/pr-1977-doctor-paths
...
fix(doctor): quote cache paths and respect release channel tags
2026-03-03 00:48:00 +09:00
b5100d99df
test(look-at): stabilize image-converter tests across platforms
2026-03-03 00:47:21 +09:00
4123148376
fix(look-at): temp dir cleanup, Windows compat, argument injection prevention
2026-03-03 00:38:47 +09:00
95fe698817
fix(model-fallback): correct transform expectations and hermetic test isolation
2026-03-03 00:38:45 +09:00
031967857f
fix(delegate-task): resolve timeout detection and config drift in sync poller
2026-03-03 00:38:22 +09:00
c80a74c5f4
fix(model-resolution): normalize model format and remove dead config flag
2026-03-03 00:38:20 +09:00
3d66a30406
fix(doctor): quote paths and respect version channels in fix messages
2026-03-03 00:38:19 +09:00
cf40ca5553
fix(hooks): ensure notification fallback on terminal-notifier failure
2026-03-03 00:38:17 +09:00
3363f0c63a
fix(test): resolve test failure in PR #2021 fixes
2026-03-03 00:27:53 +09:00
c084cc3f26
fix(dispatch): resolve plugin namespace parsing, template substitution, and discovery duplication
2026-03-03 00:14:01 +09:00
f383d7abb5
Revert "Merge pull request #1951 from edxeth/feat/custom-agents"
...
This reverts commit 47e300b17e243ce1b7e8
2026-03-02 23:55:48 +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
47e300b17e
Merge pull request #1951 from edxeth/feat/custom-agents
...
feat(config): make custom agents first-class for planning and delegation
2026-03-02 23:28:57 +09:00
JimMoen