Semble
ActiveDescription
Fast and accurate code search engine designed for AI agents. Uses semantic indexing instead of grep+read, reducing token consumption by roughly 98% while delivering higher-quality retrieval for code navigation tasks.
Key Features
- Lightning-fast semantic search — embedding-based code retrieval that reduces token usage by ~98% vs grep+read
- MCP server — exposes Model Context Protocol endpoints so agents can call it directly
- High recall — outperforms traditional text matching on multiple code-retrieval benchmarks
- Lightweight deployment — starts with a single binary/package, no heavy infrastructure required
- Semantic deduplication — automatically identifies similar functions and duplicated code snippets
- Agent-friendly — output format designed for LLM consumption
Use Cases
💡 Plugging high-speed code search into Claude Code, Cursor, Codex, and other coding agents
💡 Rapidly locating relevant functions and definitions in large monorepos
💡 Replacing grep+read workflows to dramatically cut token consumption
💡 Letting any MCP-compatible agent call Semble for code retrieval
💡 Providing semantic code completion and navigation in IDE plugins
Strengths & Limitations
✅ Strengths
- • Actively maintained, recent updates
- • High community interest (5.9k stars)
- • Permissive open-source license (MIT)
- • Responsive to issues, low backlog
Categories
Quick Start
pip install semble
semble index /path/to/repo
semble search "function that handles user authentication"
# Or start the MCP server:
semble serve --mcp