Commit Graph

3393 Commits

Author SHA1 Message Date
acamq cc833b182f fix(docs): remove corrupted text and duplicate entries in AGENTS.md
- Remove accidental '7ZB|' keystroke insertion on line 7
- Remove duplicate schema tree entries (start-work.ts and internal/permission.ts)
2026-02-27 13:45:35 -07:00
acamq d38d98ba02 feat(start-work): add auto_commit config option
Add start_work.auto_commit configuration option to allow users to
disable the automatic commit step in the /start-work workflow.

When auto_commit is false:
- STEP 8: COMMIT ATOMIC UNIT is removed from orchestrator reminder
- STEP 9: PROCEED TO NEXT TASK becomes STEP 8

Resolves #2197
2026-02-27 13:38:52 -07:00
acamq 2f4d9316c0 fix(hephaestus): add tool call format instructions to prevent malformed output
GPT-5.x Codex models occasionally hallucinate malformed tool calls like:
  assistant to=functions.XXX <garbled_unicode>json\n{...}

This is a model-level issue where the model outputs tool calls as text
instead of using the native tool calling mechanism. Add explicit prompt
instructions telling the model to use the tool call interface.

Related: #2190
2026-02-27 12:58:05 -07:00
acamq 1ab02b4f44 fix(skill-mcp): clarify builtin MCP error hint 2026-02-27 10:03:20 -07:00
David Hardy 9464d641df fix: initialize config context in plugin runtime to prevent warnings
The auto-update checker hook calls getConfigDir() which requires the
config context to be initialized via initConfigContext(). When running
in the OpenCode TUI plugin runtime (vs CLI), this initialization never
happened, causing the warning:

"getConfigContext() called before initConfigContext(); defaulting to CLI paths."

This warning would appear when opening folders containing .mulch or .beads
directories because the lifecycle plugins triggered the auto-update checker.

Fix: Call initConfigContext("opencode", null) at plugin startup to ensure
the config context is properly initialized for all hooks and utilities.

Fixes upstream issue where TUI users see spurious bun install warnings.
2026-02-27 15:19:21 +00:00
Nguyen Khac Trung Kien ca2e046eb5 Merge pull request #2161 from acamq/fix/duplicate-agent-entries 2026-02-27 18:35:59 +07:00
Kenny e18aa71d5c Merge pull request #2153 from devxoul/fix/docs-master-to-dev-branch
replace master branch references with dev
2026-02-27 19:33:48 +08:00
Kenny da2f665f51 Merge pull request #2160 from dwnmf/add-russian-readme
Add Russian README
2026-02-27 19:32:14 +08:00
github-actions[bot] a6324ff1fe @mertyldrm has signed the CLA in code-yeongyu/oh-my-opencode#2184 2026-02-27 10:53:16 +00:00
Mert Yıldırım 6c3b13aa27 fix: remove console.warn that leaks into TUI textbox
getConfigContext() emitted a console.warn when called before
initConfigContext() completed. Since initConfigContext runs async
(spawns opencode --version subprocess), other modules calling
getConfigDir/getConfigJson could trigger this warning during startup.

The fallback behavior is intentional and safe (defaults to standard
CLI paths), but console.warn writes to stderr which the TUI captures,
causing the warning to render inside the user's textbox.

Fixes #2183
2026-02-27 13:51:51 +03:00
github-actions[bot] 21d796b7d9 @YLRong has signed the CLA in code-yeongyu/oh-my-opencode#2176 2026-02-27 09:06:59 +00:00
YLRong a9574d3e95 fix: remove misleading hint from replace pos-only description
The hint '(MOST COMMON for single-line edits)' misleads agents into
thinking pos-only replace is the default behavior. When agents want
to replace multiple lines but only specify pos without end, the tool
only replaces one line, causing duplicate code from retained lines.
2026-02-27 17:06:40 +08:00
Lynricsy a80fd3747c feat(delegate-task): ⚙️ make sync subagent timeout configurable via syncPollTimeoutMs
Allow users to set `background_task.syncPollTimeoutMs` in config to override
the default 10-minute sync subagent timeout. Affects sync task, sync continuation,
and unstable agent task paths. Minimum value: 60000ms (1 minute).

Co-authored-by: Wine Fox <fox@ling.plus>
2026-02-27 16:17:39 +08:00
1noilimrev 7130ae269b refactor(test): extract shared mock helper and add try-finally for env cleanup 2026-02-27 15:30:13 +09:00
YeonGyu-Kim 50573da6c5 fix(atlas): use start-work session agent for continuation gating
Prefer the in-memory session agent set by /start-work when validating idle continuation eligibility, so stale message storage agent values do not block boulder continuation.
2026-02-27 15:01:44 +09:00
github-actions[bot] a34ac6c119 @1noilimrev has signed the CLA in code-yeongyu/oh-my-opencode#2166 2026-02-27 05:53:45 +00:00
1noilimrev f386ae265f test(hooks): add darwin notification backend selection tests 2026-02-27 14:48:34 +09:00
1noilimrev f4bf6fe4dc fix(hooks): use terminal-notifier for macOS notification click-to-focus 2026-02-27 14:39:07 +09:00
LYA_CAP_OCEAN 59d9836f9e Fix markdown blockquote structure in Russian README 2026-02-27 01:59:29 +03:00
acamq 44ea23fe7a fix(agents): remove duplicate remapped agent keys 2026-02-26 15:58:02 -07:00
github-actions[bot] a5a9184ca7 @dwnmf has signed the CLA in code-yeongyu/oh-my-opencode#2160 2026-02-26 22:51:52 +00:00
LYA_CAP_OCEAN e16cb692cb Add Russian README 2026-02-27 01:49:51 +03:00
edxeth 7729aa520b refactor(schema): dedupe custom agent override with ref 2026-02-26 21:39:04 +01:00
edxeth 42b1e6985c fix(config): avoid conflicting typo and migration guidance 2026-02-26 21:28:00 +01:00
edxeth e6f7abc523 fix(custom-agents): align planner catalog and schema validation 2026-02-26 21:14:00 +01:00
edxeth 6f8ec9b49b docs(config): fix custom_agents examples 2026-02-26 20:55:58 +01:00
YeonGyu-Kim f4f2278cb2 fix(ci): pre-download baseline compile targets to avoid Bun extraction failures
Bun's internal download of baseline compile targets from npm registry
consistently fails on Windows CI runners (ExtractionFailed error).
Pre-download the baseline binary via curl into Bun's cache directory
so the compile step finds it already cached and skips the download.

Also makes publish job resilient with if: always() so one failed
platform doesn't block publishing all other successful platforms.
2026-02-27 04:43:29 +09:00
YeonGyu-Kim 9b24ffa202 fix(ci): make platform publish resilient to individual build failures
publish job now runs with if: always() && !cancelled(), and gates
each publish step on download.outcome == 'success'. One flaky target
(e.g. windows-x64-baseline) no longer blocks all other platforms.
2026-02-27 04:39:05 +09:00
github-actions[bot] 54ed45059b release: v3.9.0 2026-02-26 19:30:38 +00:00
YeonGyu-Kim e548243378 fix(hephaestus): remove auto-commit policy to prevent surprise commits
The auto-commit section instructed Hephaestus to automatically commit after
implementation work. Users who didn't know about this behavior would get
surprise commits — a trust-breaking behavioral change flagged by 5 Oracle
reviews as the sole publish blocker for 3.9.0.
2026-02-27 04:27:07 +09:00
edxeth b34f0a5068 docs(config): document custom_agents behavior and delegation flow 2026-02-26 20:01:53 +01:00
YeonGyu-Kim ccce641ccd Merge pull request #2159 from code-yeongyu/fix/ralph-loop-completion-scoping
fix(ralph-loop): scope completion detection to messages since loop start
2026-02-27 03:23:46 +09:00
YeonGyu-Kim 9cf140ddba Merge pull request #2158 from code-yeongyu/fix/hashline-diff-format-compat
test(hashline-edit): verify diff format compatibility with OpenCode UI
2026-02-27 03:23:43 +09:00
YeonGyu-Kim 1f79dfa544 Merge pull request #2157 from code-yeongyu/fix/token-limiter-safe-truncation
fix(token-limiter): truncate at newline boundaries instead of raw slice
2026-02-27 03:23:40 +09:00
YeonGyu-Kim a0954268a4 Merge pull request #2156 from code-yeongyu/fix/background-pending-notif-leak
fix(background-agent): clean pendingNotifications on session.deleted
2026-02-27 03:23:38 +09:00
YeonGyu-Kim 973952af30 Merge pull request #2155 from code-yeongyu/fix/ultrawork-thinking-db-write
fix(ultrawork-db): write $.thinking alongside $.variant in deferred model override
2026-02-27 03:23:30 +09:00
YeonGyu-Kim 5f7bdf6cdd test(ralph-loop): harden completion detector PluginInput mock 2026-02-27 03:12:22 +09:00
YeonGyu-Kim cf3aed301c fix(ralph-loop): persist session message count at loop start 2026-02-27 03:08:30 +09:00
YeonGyu-Kim e034930194 fix(ralph-loop): scope completion detection to messages since loop start 2026-02-27 03:05:14 +09:00
YeonGyu-Kim fcd8e1080b test(hashline-edit): add diff format compatibility tests 2026-02-27 03:02:49 +09:00
YeonGyu-Kim 44b8ec750f fix(token-limiter): truncate at newline boundaries instead of raw slice 2026-02-27 03:02:04 +09:00
YeonGyu-Kim 16672bae16 fix(background-agent): clean pendingNotifications on session.deleted 2026-02-27 03:00:39 +09:00
YeonGyu-Kim a5b6681390 fix(ultrawork-db): write $.thinking alongside $.variant in deferred model override 2026-02-27 02:59:22 +09:00
edxeth 1fab8db9fe Merge remote-tracking branch 'origin/dev' into feat/custom-agents
# Conflicts:
#	src/agents/utils.test.ts
#	src/plugin-handlers/agent-config-handler.ts
2026-02-26 18:53:29 +01:00
YeonGyu-Kim 485ef2476f Merge pull request #2154 from minpeter/feat/hashline-benchmark
fix(hashline-edit): harden deduplication, validation, and add benchmark suite
2026-02-27 02:14:12 +09:00
minpeter a2b2cb7e87 fix(benchmarks): address review feedback on error handling and validation
- headless.ts: emit error field on tool_result when output starts with Error:
- test-multi-model.ts: errored/timed-out models now shown as RED and exit(1)
- test-multi-model.ts: validate --timeout arg (reject NaN/negative)
- test-edge-cases.ts: use exact match instead of trim() for whitespace test
- test-edge-cases.ts: skip file pre-creation for create-via-append test

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

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-02-27 01:44:51 +09:00
minpeter 3cff4c1bce feat(benchmarks): add hashline-edit test suites (46 tests)
Ported from code-editing-agent benchmark:
- test-edit-ops.ts: 21 basic edit operations (replace, append, prepend, delete, batch, range)
- test-edge-cases.ts: 25 edge cases (unicode, long lines, whitespace, special chars, file creation)
- test-multi-model.ts: multi-model comparison runner

Verified 21/21 + 25/25 (100%) with Minimax M2.5 via FriendliAI.

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

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-02-27 01:37:49 +09:00
minpeter a71a97c2ea feat(benchmarks): add hashline-edit benchmark agent and deps
Standalone headless agent using Vercel AI SDK v6 with FriendliAI provider.
Imports hashline-edit pure functions directly from src/ for benchmarking
the edit tool against LLMs (Minimax M2.5 via FriendliAI).

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

Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
2026-02-27 01:37:40 +09:00
minpeter 26004f93ec Fix hashline-edit deduplication and validation
- Canonicalize anchors in dedupe keys to handle whitespace variants
- Make lines field required in edit operations
- Only allow unanchored append/prepend to create missing files
- Reorder delete/rename validation to prevent edge cases
- Add allow_non_gpt_model and max_prompt_tokens to config schema
  ```
2026-02-27 01:37:19 +09:00
ismeth 05f1e1fedb fix(agent-usage-reminder): skip reminders for non-orchestrator subagents 2026-02-26 17:05:33 +01:00