Mcp is a Model Context Protocol (MCP) server: The grounded data layer for any LLM: governed SQL, metrics, lineage and catalog over your data. In practice that means any MCP-compatible AI assistant can call Mcp's tools directly — the model decides when to use them in a conversation or agent run.
You can run Mcp two ways: connect to the hosted endpoint at https://mcp.matih.ai/api/v1/mcp with no local install, or run it on your own machine via the npm package @matihlabs/mcp. Configuration happens through environment variables — you'll need MATIH_MCP_TOKEN. The current release is v0.2.1, with source at matih-labs/matih-mcp on GitHub.
Mcp is listed under Database on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain Mcp for matih.ai, claim this listing to verify ownership, earn the Verified badge, and keep the details current.
