MCP Local RAG is a local semantic search tool for developers that allows searching through documents without sending data to external APIs. It supports keyword boosting for exact technical terms and operates entirely offline after initial setup.
From the registry: Easy-to-setup local RAG server with minimal configuration

Please install the `mcp-local-rag` MCP server into my current AI client (that's you).
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","mcp-local-rag"]
- required environment variables:
- `BASE_DIR`: Document root directory (security boundary) (example: `/path/to/your/documents`)
- optional environment variables:
- `DB_PATH`: Vector database location (example: `./lancedb/`)
- `CACHE_DIR`: Model cache directory (example: `./models/`)
- `MODEL_NAME`: HuggingFace embedding model ID (example: `Xenova/all-MiniLM-L6-v2`)
- `MAX_FILE_SIZE`: Maximum file size in bytes (example: `104857600`)
- `RAG_HYBRID_WEIGHT`: Keyword boost factor (0=semantic only) (example: `0.6`)
Note: Fully local RAG with Transformers.js and LanceDB. First run downloads ~90MB model.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.BASE_DIRrequiredThe folder where documents to be searched are located.AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
MCP Toolbox for Databases enables your agent to connect to your database.
MCP Toolbox for Databases enables your agent to connect to your database.