A curated, searchable directory of Model Context Protocol (MCP) servers — filtered by category, use case, and complexity. Find the right integration for your Claude setup in under 60 seconds.
By Zach Bailey
The Model Context Protocol (MCP) is an open-source standard that allows AI models like Claude to interact directly with local data, remote APIs, and development tools. MCP servers act as the bridge, giving the AI "tools" to perform actions like searching GitHub, reading local files, or querying databases.
No servers found matching your criteria. Try another search terms.
By default, LLMs are isolated from your live data. MCP changes this. This directory helps you find the specific bridge you need to let Claude read your local codebase, search your Slack history, or even execute terminal commands safely.
Most MCP servers are installed by adding a configuration block to your Claude Desktop config file (`claude_desktop_config.json`). You'll typically need to provide an API key or a local file path depending on the server.
Because MCP servers run locally on your machine (or on servers you control), you have complete oversight. You only grant access to specific directories or APIs that you explicitly configure.
Yes. Anthropic provides a Python and TypeScript SDK that makes it easy to wrap any API or data source in an MCP-compliant interface. Check our guides below for a step-by-step tutorial.