メニュー
Personal knowledge management expert for the brain CLI. Use when users want to capture notes, search their knowledge base, or manage their second brain.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | brain-expert |
| description | Personal knowledge management expert for the brain CLI. Use when users want to capture notes, search their knowledge base, or manage their second brain. |
You are an expert on the brain CLI - a personal knowledge management tool with hierarchical Markdown files, SQLite storage, and semantic search.
| Command | Description |
|---|---|
brain init [path] | Initialize brain directory |
brain add "text" | Add timestamped note to today's file |
brain add -f file.md "text" | Add to specific file |
brain add --ref /path "text" | Add note referencing external file |
brain new "path/name" | Create new entry, opens in editor |
brain list | Show recent entries |
brain list --tree | Show directory structure |
brain show <path> | Display entry content |
brain edit <path> | Open entry in editor |
brain delete <path> | Delete entry (file + database) |
brain rm <path> | Alias for delete |
brain sync | Sync files to database |
brain sync --force | Re-embed everything |
brain search "query" | Semantic search (default) |
brain search --exact "term" | Full-text search (FTS5) |
brain config show | Display configuration |
brain todo add "text" | Create a new todo |
brain todo list | List open todos |
brain todo done <id> | Mark todo as complete |
brain todo cancel <id> | Cancel a todo |
brain todo edit <id> | Edit todo in editor |
brain todo rm <id> | Delete todo permanently |
brain cron install | Install auto-sync cron job |
brain cron status | Check auto-sync status |
brain cron remove | Remove auto-sync cron job |
Brain organizes files hierarchically:
~/Documents/brain/
├── 2026/
│ └── 01/
│ ├── 22.md # Daily journal (timestamped entries)
│ └── 23.md
├── projects/
│ ├── acme.md # Named entries (by topic)
│ └── startup-ideas.md
├── notes/
│ └── meeting-notes.md
└── .brain.db # SQLite database (gitignored)
Auto-created when using brain add:
[2026-01-23-143022]
First thought of the day
[2026-01-23-153045]
Another thought with more context
Created with brain new:
# Project Title
Content organized however you like.
Can include todos: - [ ] Task here
Uses OpenAI embeddings for meaning-based search:
brain search "database optimization strategies"
Returns results ranked by semantic similarity.
Uses SQLite FTS5 for literal text matching:
brain search --exact "PostgreSQL"
Before searching, ensure database is synced:
brain sync
The sync process:
.md filesUse brain sync --force to re-embed everything.
Config stored at ~/.brain.json:
{
"path": "/Users/taras/Documents/brain",
"editor": "code",
"embeddingModel": "text-embedding-3-small"
}
Manage with:
brain config show - view configbrain config set editor vim - update valuebrain add "Idea: could use SQLite for local caching"
brain add --ref ./src/api/auth.ts "Need to add rate limiting here"
brain search "authentication patterns"
brain new "projects/new-feature"
# Opens editor with # New Feature header
brain delete 2026/01/22 # Delete with confirmation
brain rm notes/old-project # Alias, same behavior
brain rm --force notes/temp # Skip confirmation
brain rm --db-only notes/draft # Keep file, remove from database
# Add todos with options
brain todo add "Review PR"
brain todo add -p myproject "Ship feature"
brain todo add -d tomorrow "Deploy to prod"
brain todo add -p work -d "next week" "Plan sprint"
# List and filter
brain todo list # Open todos
brain todo list --all # Include done/cancelled
brain todo list -p myproject # Filter by project
# Complete and manage
brain todo done 1 2 3 # Mark multiple as done
brain todo cancel 1 # Cancel a todo
brain todo edit 1 # Edit in $EDITOR
brain todo rm 1 # Delete permanently
# Set up background sync (runs every N minutes)
brain cron install # Default: 5 minutes
brain cron install --interval 15 # Custom interval
# Check and manage
brain cron status # "Active (every 5 minutes)"
brain cron remove # Stop auto-sync
| Issue | Solution |
|---|---|
| Search returns nothing | Run brain sync first |
| Embedding errors | Check OPENAI_API_KEY is set |
| Command not found | Run bun link in brain directory |
| Wrong brain path | Check brain config show |
File review tool — launch GUI, process comments, or install. Use when user mentions file-review, reviewing files, leaving comments, or processing review comments.
Execute a single DAG step as an autonomous background sub-agent. Sibling of phase-running for DAG plans produced by v-planning. Reads a step-<n>.md file directly, atomically claims it via frontmatter status, runs the three-bucket Success Criteria, and reports back. Spawned by v-implementing or by /run-step.
Parallel DAG-plan implementation skill. Reads a v-planning plan directory (root.md + step-<n>.md files), topologically schedules ready steps, and fans them out as parallel sub-agents. Use whenever the user invokes /v-implement, points at a plan directory produced by /v-plan, or asks to "run the parallel plan", "implement the DAG", or "fan out the steps" — even without those exact words. For linear plans (single .md file), use `implementing` instead.
Vertical / parallel implementation planning skill. Creates DAG-structured plan directories where each step is an independent, QA-able vertical slice that sub-agents can pick up and implement in parallel. Use whenever the user wants a plan that fans out (multiple independent features), invokes /v-plan, or asks for a "parallel plan", "DAG plan", "vertical plan", or "plan that can be parallelized" — even if they don't say those exact words. Prefer the linear `planning` skill for strictly sequential work.
Interactive exploration of ideas through Socratic Q&A. Produces progressive documents that serve as lightweight pre-PRDs feeding into research.
Plan implementation skill. Executes approved technical plans phase by phase with verification checkpoints.