Commit Graph

161 Commits

Author SHA1 Message Date
YeonGyu-Kim e147be7ed4 feat: update comment-checker to v0.5.0
BREAKING CHANGE: Docstrings are now detected as code smell.
See: https://github.com/code-yeongyu/go-claude-code-comment-checker/releases/tag/v0.5.0

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 124c3b3e8f docs: fix documentation inconsistencies identified by Oracle
- Add source path annotations to Command/Skill/Agent/MCP Loaders
- Add Session State feature documentation
- MCP Loader paths verified (match loader.ts: ~/.claude/.mcp.json, ./.mcp.json, ./.claude/.mcp.json)

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 5678e0bac6 docs(readme): document new claude-code feature loaders
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 207435450c refactor(index): integrate session-state module and remove local variables
- Add imports: agent-loader, mcp-loader, session-state, logger
- Remove local session variables (mainSessionID, currentSessionID, currentSessionTitle)
- Use setter/getter functions from session-state module
- Add agent loading in config hook (loadUserAgents, loadProjectAgents)
- Add MCP loading in config hook (loadMcpConfigs)
- Replace console.error with logger

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 376bf363af feat(features): add claude-code-agent-loader, mcp-loader, session-state 2025-12-09 16:59:22 +09:00
YeonGyu-Kim c7a65af475 refactor(features): rename command-loader and skill-loader with claude-code prefix 2025-12-09 16:59:22 +09:00
YeonGyu-Kim 8e7447deee feat(shared): add file-based logger utility 2025-12-09 16:59:22 +09:00
YeonGyu-Kim 15a748b817 docs: add missing hooks and features to README
Add 4 missing hooks documentation:
- Think Mode: auto-detect deep thinking requests
- Anthropic Auto Compact: auto-compact context
- Empty Task Response Detector: handle empty responses
- Grep Output Truncator: prevent output overflow

Add 2 missing features documentation:
- Command Loader: load commands from multiple paths
- Skill Loader: load skills as executable commands

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim c0e0dc1f95 feat: integrate command/skill loaders and think-mode hook in main entry
- Add loadCommands() and loadSkills() to config
- Register think-mode hook for UserPromptSubmit event

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 7059407cbc feat(hooks): export createThinkModeHook from index
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 589cf60252 feat(hooks): add think-mode hook for automatic model switching
Detects thinking keywords (ultrathink, deepthink, etc.) and switches
to thinking-capable models automatically.

Supports model patterns:
- claude-sonnet-4-0 -> claude-sonnet-4-0-max-thinking
- claude-sonnet-4-20250514 -> claude-sonnet-4-20250514-max-thinking

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim e5cdaa5192 feat(tools): export slashcommand and skill tools from index
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 340eb30147 feat(tools): add skill tool for invoking skills in conversation
Provides 'skill' tool that invokes skills loaded by skill-loader.
Skills expand into detailed instructions when invoked.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim e72b927ccb feat(tools): add slashcommand tool for executing slash commands
Provides 'slashcommand' tool that executes commands loaded by command-loader.
Handles shell injection and file reference resolution.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 3c6ffe5d9c feat(skill-loader): add skill loader that converts skills to commands
Skills are loaded from:
- ~/.claude/skills/ (user scope)
- .claude/skills/ (project scope)

Each skill directory contains SKILL.md with frontmatter metadata.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 938a3709e1 feat(command-loader): add slash command loader from multiple paths
Load commands from 4 directory scopes:
- .opencode/command/ (opencode-project)
- .claude/commands/ (project)
- ~/.config/opencode/command/ (opencode)
- ~/.claude/commands/ (user)

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
YeonGyu-Kim 47f218e33f feat(shared): add shared utilities for command and skill loading
- frontmatter.ts: YAML frontmatter parser
- file-reference-resolver.ts: resolve @file references in markdown
- command-executor.ts: execute shell commands in markdown
- model-sanitizer.ts: sanitize model names for OpenCode compatibility

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 16:59:22 +09:00
github-actions[bot] e07a25baa4 release: v0.1.31 2025-12-09 05:42:03 +00:00
YeonGyu-Kim 08ede0a28d deps: bump @code-yeongyu/comment-checker to 0.4.4
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 14:41:07 +09:00
github-actions[bot] a711d58289 release: v0.1.30 2025-12-09 02:50:19 +00:00
YeonGyu-Kim 431ec14991 docs: update notepad with cleanup task logs
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 11:49:17 +09:00
YeonGyu-Kim 62cae8114d refactor(comment-checker): simplify binary path resolution and add separator warning
- Remove platform-specific package lookup logic
- Remove homebrew path resolution
- Add code smell warning for comment separators

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 11:49:11 +09:00
YeonGyu-Kim e6eafe267a refactor(ast-grep): remove NAPI-based tools
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 11:49:00 +09:00
YeonGyu-Kim e4ef832405 feat(hooks): add anthropic-auto-compact hook
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 11:48:53 +09:00
YeonGyu-Kim ef6d67645e refactor(hooks): remove pulse-monitor hook
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 11:48:46 +09:00
YeonGyu-Kim 227d93f106 docs(readme-ko): clarify nested AGENTS.md injection behavior with example
Added directory tree example showing how multiple AGENTS.md files are
collected and injected in hierarchical order when reading nested files.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:38:05 +09:00
github-actions[bot] edff922afb release: v0.1.29 2025-12-09 01:37:46 +00:00
YeonGyu-Kim 45bdcf3580 docs(readme): clarify nested AGENTS.md injection behavior with example
Added directory tree example showing how multiple AGENTS.md files are
collected and injected in hierarchical order when reading nested files.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:36:34 +09:00
YeonGyu-Kim b07dd22093 fix(pulse-monitor): reset heartbeat after tool execution to prevent false positives
Tools can take arbitrary time, so we need a fresh baseline after execution.
Previously, lastHeartbeat wasn't updated after tool.execute.after, causing
stalled detection to trigger immediately after long-running tools.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:36:34 +09:00
YeonGyu-Kim c7d29fea48 refactor(mcp): remove unused builtinMcps export
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:36:34 +09:00
YeonGyu-Kim 55675497a5 refactor(session-recovery): remove unused ThinkingPart interface and fallbackRevertStrategy function
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:36:34 +09:00
YeonGyu-Kim ae2d347d81 refactor(lsp): remove unused formatWorkspaceEdit import
🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-09 10:36:34 +09:00
github-actions[bot] 2683de825a release: v0.1.28 2025-12-08 08:52:28 +00:00
YeonGyu-Kim 0b5c8250ca fix(pulse-monitor): prevent false positive stalled detection after tool execution
Remove forced monitoring restart in tool.execute.after to avoid false positive
stalled session detection when LLM legitimately completes response after tool run.
Monitoring now resumes naturally on next session/message event.

🤖 GENERATED WITH ASSISTANCE OF [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
2025-12-08 17:51:06 +09:00
github-actions[bot] 66fcd8570b release: v0.1.27 2025-12-08 08:01:03 +00:00
YeonGyu-Kim 5cd3f0cbf2 docs: add Directory AGENTS.md Injector hook documentation
🤖 GENERATED BY [OPENCODE](https://opencode.ai/)
2025-12-08 17:00:02 +09:00
YeonGyu-Kim 9a9512b705 test: add test directory with sample files 2025-12-08 17:00:02 +09:00
YeonGyu-Kim 6ece7476ef feat(hooks): add empty-task-response-detector hook 2025-12-08 17:00:02 +09:00
YeonGyu-Kim 9ed23d4037 feat(hooks): implement directory-agents-injector hook 2025-12-08 17:00:02 +09:00
YeonGyu-Kim 79b791117a fix(session-recovery): improve error message extraction 2025-12-08 17:00:02 +09:00
YeonGyu-Kim 4e328a937c feat(hooks): integrate directory-agents-injector hook into plugin pipeline
- Add directoryAgentsInjector to plugin event handlers
- Wire up tool.execute.after hook for directory agents injection
- Fix: Format src/index.ts with consistent semicolon style
2025-12-08 17:00:02 +09:00
YeonGyu-Kim a500f0c9ad export(hooks): add directory-agents-injector hook to public API
- Export createDirectoryAgentsInjectorHook from hooks index
- Fix: Formatting (add semicolons to match code style)
2025-12-08 17:00:02 +09:00
YeonGyu-Kim 16806da615 refactor(session-recovery): process entire message history for empty/thinking block recovery
- Scan all non-final assistant messages for empty content, orphan thinking blocks, and disabled thinking
- Add storage utility functions: findMessagesWithThinkingBlocks, findMessagesWithOrphanThinking, stripThinkingParts, prependThinkingPart
- Fix: Previously only processed single failed message, now handles multiple broken messages in history
- Improve: Use filesystem-based recovery instead of unreliable SDK APIs
2025-12-08 17:00:02 +09:00
YeonGyu-Kim c5f651c0a9 refactor(hooks): remove grep-blocker (grep tool now overrides built-in)
The grep tool now properly overrides OpenCode's built-in grep,
making the blocker hook unnecessary.

Generated by [OpenCode](https://opencode.ai/)
2025-12-08 17:00:02 +09:00
YeonGyu-Kim ed3d7a55f4 feat(tools): add glob tool with timeout protection
- Override OpenCode's built-in glob with 60s timeout
- Kill process on expiration to prevent indefinite hanging
- Reuse grep's CLI resolver for ripgrep detection

Generated by [OpenCode](https://opencode.ai/)
2025-12-08 17:00:02 +09:00
YeonGyu-Kim b77dd2fcdf refactor(tools): rename safe-grep to grep with override capability 2025-12-08 17:00:02 +09:00
YeonGyu-Kim 64b3564760 refactor(session-recovery): extract storage utilities to separate module
Split session-recovery.ts into modular structure:
- types.ts: SDK-aligned type definitions
- constants.ts: storage paths and part type sets
- storage.ts: reusable read/write operations
- index.ts: main recovery hook logic
2025-12-08 17:00:02 +09:00
YeonGyu-Kim 0df7e9b10b fix(session-recovery): recognize 'tool' type as valid content
OpenCode storage uses 'tool' type for tool calls, but the hasContent
check only recognized 'tool_use' (Anthropic API format). This caused
messages with tool calls to be incorrectly identified as empty.
2025-12-08 17:00:02 +09:00
YeonGyu-Kim aa35f2eab6 fix(session-recovery): detect empty messages with zero parts
Previously, findEmptyContentMessageFromStorage only detected messages
with parts.length > 0 that had no content. This missed the case where
a message has zero parts entirely, causing infinite recovery loops.
2025-12-08 17:00:02 +09:00
github-actions[bot] 64db980803 release: v0.1.26 2025-12-08 00:41:34 +00:00