Commit Graph

429 Commits

Author SHA1 Message Date
YeonGyu-Kim 2d80b76b30 Merge branch 'dev' into fix/1681-oracle-json-parse 2026-02-17 01:54:03 +09:00
YeonGyu-Kim bdbc111384 Merge pull request #1896 from code-yeongyu/fix/1283-review-code-silent-fail
fix: report silent subagent delegation failures (#1283)
2026-02-17 01:53:56 +09:00
YeonGyu-Kim 12aae86dec Merge pull request #1895 from code-yeongyu/fix/1718-windows-subagent-dir
fix: use correct project directory for Windows subagents (#1718)
2026-02-17 01:53:43 +09:00
YeonGyu-Kim 3a4325a800 Merge pull request #1884 from code-yeongyu/feat/hashline-edit
feat: port hashline edit tool from oh-my-pi
2026-02-17 01:35:22 +09:00
YeonGyu-Kim f04045b335 Merge pull request #1708 from jsl9208/fix/ast-grep-replace-silent-noop
fix(ast-grep): fix ast_grep_replace silent write failure
2026-02-17 01:34:41 +09:00
YeonGyu-Kim 04ee341399 Merge pull request #1729 from potb/fix/1716-disabled-agents-call-omo
fix(call-omo-agent): enforce disabled_agents config
2026-02-17 01:34:38 +09:00
YeonGyu-Kim 2f08c1d316 fix: use correct project directory for Windows subagents (#1718) 2026-02-17 01:29:25 +09:00
YeonGyu-Kim f4fc2fbbfd fix: resolve JSON parse error in Oracle after promptAsync refactor (#1681) 2026-02-17 01:29:17 +09:00
YeonGyu-Kim 41c57df6f9 fix: report errors instead of silent catch in subagent-resolver (#1283) 2026-02-17 01:26:58 +09:00
YeonGyu-Kim bcf0a4998d test(session-manager): fix storage tests by mocking message-dir dependency 2026-02-17 00:03:10 +09:00
YeonGyu-Kim 1345c0719d test: skip 4 flaky session-manager tests (test order dependency) 2026-02-17 00:03:10 +09:00
YeonGyu-Kim eb6be8ac7b fix(hashline): address Cubic review comments
- P2: Change replace edit sorting from POSITIVE_INFINITY to NEGATIVE_INFINITY
  so replace edits run LAST after line-based edits, preventing line number
  shifts that would invalidate subsequent anchors

- P3: Update tool description from SHA-256 to xxHash32 to match actual
  implementation in hash-computation.ts
2026-02-17 00:03:10 +09:00
YeonGyu-Kim 9bd21c8a67 feat(hashline): port hashline edit tool from oh-my-pi
This PR ports the hashline edit tool from oh-my-pi to oh-my-opencode as an experimental feature.

## Features
- New experimental.hashline_edit config flag
- hashline_edit tool with 4 operations: set_line, replace_lines, insert_after, replace
- Hash-based line anchors for safe concurrent editing
- Edit tool disabler for non-OpenAI providers
- Read output enhancer with LINE:HASH prefixes
- Provider state tracking module

## Technical Details
- xxHash32-based 2-char hex hashes
- Bottom-up edit application to prevent index shifting
- OpenAI provider exemption (uses native apply_patch)
- 90 tests covering all operations and edge cases
- All files under 200 LOC limit

## Files Added/Modified
- src/tools/hashline-edit/ (7 files, ~400 LOC)
- src/hooks/hashline-edit-disabler/ (4 files, ~200 LOC)
- src/hooks/hashline-read-enhancer/ (3 files, ~400 LOC)
- src/features/hashline-provider-state.ts (13 LOC)
- src/config/schema/experimental.ts (hashline_edit flag)
- src/config/schema/hooks.ts (2 new hook names)
- src/plugin/tool-registry.ts (conditional registration)
- src/plugin/chat-params.ts (provider state tracking)
- src/tools/index.ts (export)
- src/hooks/index.ts (exports)
2026-02-17 00:03:10 +09:00
YeonGyu-Kim 3e90ec6f93 feat(config): add experimental.hashline_edit flag and provider state module 2026-02-17 00:03:10 +09:00
YeonGyu-Kim ba770dc9ec test: align load_skills error assertions in delegate-task 2026-02-16 22:59:52 +09:00
YeonGyu-Kim 41286affff fix: force include_thinking and include_tool_results for running background tasks 2026-02-16 22:47:51 +09:00
YeonGyu-Kim cabd1e9696 feat: auto-enable full_session, thinking, and tool_results for running background tasks 2026-02-16 22:37:27 +09:00
YeonGyu-Kim 9cc941e0cf refactor: remove dead file task-id-validator.ts and unused isModelAvailable from model-name-matcher 2026-02-16 22:33:44 +09:00
YeonGyu-Kim d4f469827f refactor: remove unused params/imports/types from lsp-tools, task-tools, delegate-task, skill-loader, context-window-monitor, plugin-config 2026-02-16 22:12:21 +09:00
YeonGyu-Kim 7c6da2b1f8 refactor: remove unused code from background-agent, background-task, call-omo-agent 2026-02-16 22:11:29 +09:00
YeonGyu-Kim e5be444c07 remove dead code: legacy unified task tool and its action handlers 2026-02-16 21:58:44 +09:00
YeonGyu-Kim d7b4ed7c2a clarify task tool: emphasize category/subagent_type is required, remove inline examples 2026-02-16 21:47:56 +09:00
YeonGyu-Kim 513e618bab fix: resolve category delegation and command routing with display name agent keys
Category-based delegation (task(category='quick')) was broken because
SISYPHUS_JUNIOR_AGENT sent 'sisyphus-junior' to session.prompt but
config.agent keys are now display names ('Sisyphus-Junior').

- Use getAgentDisplayName() for SISYPHUS_JUNIOR_AGENT constant
- Replace hardcoded 'sisyphus-junior' strings in tools.ts with constant
- Update background-output local constants to use display names
- Add remapCommandAgentFields() to translate command agent fields
- Add raw-key fallback in tool-config-handler agentByKey()
2026-02-16 21:32:33 +09:00
YeonGyu-Kim 49b6c108e2 Normalize agent name comparisons to handle display name keys
Hooks and tools now use getAgentConfigKey() to resolve agent names (which may
be display names like 'Atlas (Plan Executor)') to lowercase config keys
before comparison.

- session-utils: orchestrator check uses getAgentConfigKey
- atlas event-handler: boulder agent matching uses config keys
- category-skill-reminder: target agent check uses config keys
- todo-continuation-enforcer: skipAgents comparison normalized
- subagent-resolver: resolves 'metis' -> 'Metis (Plan Consultant)' for lookup
2026-02-16 20:43:09 +09:00
YeonGyu-Kim 8582fe267b fix: resolve session-manager storage test mock pollution (pre-existing CI failure) 2026-02-16 18:29:30 +09:00
YeonGyu-Kim 63e983269d refactor: consolidate duplicated Promise.all dual reads into resolveMessageContext utility 2026-02-16 18:20:27 +09:00
YeonGyu-Kim 0b10245c34 refactor: create normalizeSDKResponse helper and replace scattered patterns across 37 files 2026-02-16 18:20:19 +09:00
YeonGyu-Kim 90555f3d5e fix: propagate sessionExists SDK errors instead of swallowing them 2026-02-16 16:52:27 +09:00
YeonGyu-Kim 64000785c7 fix: handle array-shaped SDK responses in getSdkMessages & dedup getMessageDir
- getSdkMessages now handles both response.data and direct array
  responses from SDK
- Consolidated getMessageDir: storage.ts now re-exports from shared
  opencode-message-dir.ts (with path traversal guards)
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 9c2a92c235 docs: regenerate all 13 AGENTS.md files from deep codebase exploration 2026-02-16 16:13:40 +09:00
YeonGyu-Kim 052f52822c fix: address 8-domain Oracle review findings (C1, C2, M1-M4)
- C1: thinking-prepend unique part IDs per message (global PK collision)
- C2: recover-thinking-disabled-violation try/catch guard on SDK call
- M1: remove non-schema truncated/originalSize fields from SDK interfaces
- M2: messageHasContentFromSDK treats thinking-only messages as non-empty
- M3: syncAllTasksToTodos persists finalTodos + no-id rename dedup guard
- M4: AbortSignal.timeout(30s) on HTTP fetch calls in opencode-http-api

All 2739 tests pass, typecheck clean.
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 4c06c66c45 fix: address all Cubic P2 review issues
- session-utils: log SDK errors instead of silent swallow
- opencode-message-dir: fix indentation, improve error log format
- storage: use session.list for sessionExists (handles empty sessions)
- storage.test: use resetStorageClient for proper SDK client cleanup
- todo-sync: add content-based fallback for id-less todo removal
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 9c0355f04c fix(test): fix sync continuation test mock leaking across sessions
The messages() mock in 'session_id with background=false' test did not
filter by session ID, causing resolveParentContext's SDK calls for
parent-session to increment messagesCallCount. This inflated
anchorMessageCount to 4 (matching total messages), so the poll loop
could never detect new messages and always hit MAX_POLL_TIME_MS.

Fix: filter messages() mock by path.id so only target session
(ses_continue_test) increments the counter. Restore MAX_POLL_TIME_MS
from 8000 back to 2000.
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 532c222d7a fix(test): increase timeouts for CI-flaky polling tests
- runner.test.ts: waitForEventProcessorShutdown timeout 50ms → 500ms
  (50ms was consistently too tight for CI runners)
- tools.test.ts: MAX_POLL_TIME_MS 2000ms → 8000ms
  (polling timed out at ~2009ms on CI due to resource contention)
2026-02-16 16:13:40 +09:00
YeonGyu-Kim e720a6cc26 fix: make sessionExists() async with SDK verification on SQLite
sessionExists() previously returned unconditional true on SQLite,
preventing ralph-loop orphaned-session cleanup from triggering.
Now uses sdkClient.session.messages() to verify session actually
exists. Callers updated to await the async result.

Addresses Cubic review feedback on PR #1837.
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 8b9379e4e5 fix: make sessionExists() SQLite-aware for session_read tool
sessionExists() relied on JSON message directories which don't exist on
SQLite. Return true on SQLite and let readSessionMessages() handle lookup.
Also add empty-messages fallback in session_read for graceful not-found.
2026-02-16 16:13:40 +09:00
YeonGyu-Kim dd618972de feat: migrate tool callers to SDK message finders on SQLite backend 2026-02-16 16:13:40 +09:00
YeonGyu-Kim 836ad0f97d feat: migrate call-omo-agent tool callers to SDK message finders 2026-02-16 16:13:40 +09:00
YeonGyu-Kim d4b9246573 refactor(shared): unify MESSAGE_STORAGE/PART_STORAGE constants into single source
- Create src/shared/opencode-storage-paths.ts with all 4 constants
- Update 4 previous declaration sites to import from shared file
- Update additional OPENCODE_STORAGE usages for consistency
- Re-export from src/shared/index.ts
- No duplicate constant declarations remain
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 6684e5aed5 fix: prefer id matching when deleting todos (Cubic feedback)
- When deleting tasks, prefer matching by id if present

- Fall back to content matching only when todo has no id

- Prevents deleting unrelated todos with same subject
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 8310ca3d12 fix: handle deleted tasks in todo-sync (Cubic feedback)
- When task is deleted (syncTaskToTodo returns null), filter by content

- Prevents stale todos from remaining after task deletion
2026-02-16 16:13:40 +09:00
YeonGyu-Kim bf5ae81357 fix: address remaining Cubic review comments (P2 issues)
- Add content-based fallback matching for todos without ids

- Add TODO comment for exported but unused SDK functions

- Add resetStorageClient() for test isolation

- Fixes todo duplication risk on beta (SQLite backend)
2026-02-16 16:13:40 +09:00
YeonGyu-Kim fb61d6235c feat(session-manager): add version-gated SDK read path for OpenCode beta
- Add SDK client injection via setStorageClient()

- Version-gate getMainSessions(), getAllSessions(), readSessionMessages(), readSessionTodos()

- Add comprehensive tests for SDK path (beta mode)

- Maintain backward compatibility with JSON fallback
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 217b1919cc refactor(shared): unify MESSAGE_STORAGE/PART_STORAGE constants into single source
- Add src/shared/opencode-storage-paths.ts with consolidated constants

- Update imports in hook-message-injector and session-manager

- Add src/shared/opencode-storage-detection.ts with isSqliteBackend()

- Add OPENCODE_SQLITE_VERSION constant

- Export all from shared/index.ts
2026-02-16 16:13:40 +09:00
YeonGyu-Kim 1d5920235f refactor(shared): consolidate 13+ getMessageDir copies into single shared function 2026-02-16 16:13:39 +09:00
YeonGyu-Kim ed098925b5 fix(todo): make Todo id field optional for OpenCode beta compatibility
- Make id field optional in all Todo interfaces (TodoInfo, Todo, TodoItem)
- Fix null-unsafe comparisons in todo-sync.ts to handle missing ids
- Add test case for todos without id field preservation
- All tests pass and typecheck clean
2026-02-16 16:13:39 +09:00
YeonGyu-Kim 39d1ee6c30 fix: detect interrupted/error/cancelled status in unstable-agent-task polling loop
The polling loop in executeUnstableAgentTask only checked session status
and message stability, never checking if the background task itself had
been interrupted. This caused the tool call to hang until MAX_POLL_TIME_MS
(10 minutes) when a task was interrupted by prompt errors.

Add manager.getTask() check at each poll iteration to break immediately
on terminal statuses (interrupt, error, cancelled), returning a clear
failure message instead of hanging.
2026-02-16 15:56:52 +09:00
YeonGyu-Kim f3672c527e feat(writing): switch primary model to kimi k2.5, add anti-AI-slop rules to prompt 2026-02-15 14:00:03 +09:00
YeonGyu-Kim 2434e5d255 feat(visual-engineering): add variant high to gemini-3-pro, update fallback chain to gemini→glm-5→opus→kimi 2026-02-15 13:59:00 +09:00
YeonGyu-Kim 805b80bbec test(skill-tool): restore bun mocks after tests 2026-02-14 22:06:46 +09:00