| name | byterover |
| description | Knowledge management for AI agents. Use the `brv` CLI to store and retrieve project patterns, decisions, and architectural rules in .brv/context-tree. Use before work (brv query) and after implementing (brv curate). Install: npm install -g byterover-cli. |
ByteRover Knowledge Management
Use the brv CLI to manage your project's long-term memory.
Install: Optional. With npm, byterover-cli is an optional dependency of ClawAgents (so brv is available from the project's node_modules/.bin). With Python, the agent runs brv via npx byterover-cli, so Node/npx is sufficient. You can also install globally: npm install -g byterover-cli.
Knowledge is stored in .brv/context-tree/ as human-readable Markdown. No authentication needed for query/curate; login only for cloud sync.
Workflow
- Before thinking: Run
brv query to understand existing patterns.
- After implementing: Run
brv curate to save new patterns/decisions.
Commands
Query knowledge
Use when the user wants recall, your context lacks needed info, or before an action to check rules/preferences. Do not use when the info is already in context.
brv query "How is authentication implemented?"
brv query "What are the API rate limits?"
Headless mode with JSON output (for automation):
brv query "How does auth work?" --headless --format json
Curate context
Use when the user wants you to remember something, or there are meaningful decisions to persist. Do not use for transient or general knowledge.
brv curate "Auth uses JWT with 24h expiry. Tokens stored in httpOnly cookies via authMiddleware.ts"
With source files (max 5):
brv curate "Authentication middleware details" -f src/middleware/auth.ts
brv curate "JWT implementation" --files src/auth/jwt.ts --files docs/auth.md
Curate an entire folder:
brv curate --folder src/auth/
brv curate "Analyze authentication module" -d src/auth/
Project setup
Initialize ByteRover for the current project:
brv init
brv init --force
Authentication (for cloud sync)
brv login --api-key YOUR_KEY
Cloud sync (optional)
brv pull
brv pull --branch feature-branch
brv push
Status
brv status
Error handling
- "No ByteRover instance is running" → run
brv in a separate terminal first
- "Not logged in" → only needed for push/pull; run
brv login --api-key <key>
- "Maximum 5 files allowed" → use ≤5
-f flags
- "File does not exist" → verify path from project root
- "Sandbox environment detected" → run
brv outside the sandbox/IDE terminal
Best practices
- Query before starting work; curate after completing work.
- Use precise queries and concise, specific curate text.
- Attach files with
-f instead of pasting content.
- Use
-d to curate an entire folder when multiple related files changed.
- Mark outdated knowledge as OUTDATED when replacing it.
- Use
--headless --format json for automation pipelines.
Source: ByteRover on ClawHub