local-faiss-mcp
Local FAISS vector database for RAG with document ingestion, semantic search, and MCP prompts.
DatabasesWeb & Searchpython
How to use
local-faiss-mcp is an MCP server that lets your AI assistant read from and write to your database straight from a conversation.
- Best for letting an assistant inspect schemas, run queries, and summarize results.
- Works in any MCP-compatible client — Claude Desktop, Claude Code, Cursor, and others.
- Add it once to your client config; its tools then appear to the assistant automatically.
Install
Distributed as a PyPI package. Prerequisite: Python 3.10+ and uv — https://docs.astral.sh/uv/.
- 1Install the prerequisite above if you don't have it.
- 2Open your Claude Desktop config (create it if missing):
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
- macOS:
- 3Add this under
mcpServers(merge if the key already exists):{ "mcpServers": { "local-faiss-mcp": { "command": "uvx", "args": ["local-faiss-mcp"] } } } - 4Save the file and fully quit, then reopen Claude Desktop.
- 5Click the tools (🔌) icon — local-faiss-mcp's tools should now be listed.
Capabilities
No capabilities documented for this server yet.
Reviews (0)
No reviews yet. Be the first to review this server.
