Find a ready-made MCP server to connect Claude Desktop or Cursor to your database, file system, or Slack workspace.
Browse MCP servers by category to add search, finance, or calendar tools to your AI assistant without writing custom integrations.
Identify which MCP implementations are officially supported by the protocol authors versus community-contributed ones.
Evaluate the security posture of an MCP server before installing it using the README's security guidance.
MCP servers run with the same system permissions as the application that started them, review each server's code before installing and grant only needed permissions.
This repository is a curated directory of servers built on a standard called Model Context Protocol, or MCP. MCP is an open specification that defines how AI assistants can connect to external tools and data sources in a consistent way. Instead of every AI tool building its own one-off connections, MCP gives developers a shared format so that an AI can talk to a file system, a database, a web search service, or almost anything else using the same basic approach. The list itself is organized by category. There are entries for file systems, databases, version control systems, cloud storage, communication tools like Slack and email, search and web browsing, location services, finance, note-taking, social media, code development tools, and many more. Each entry links to an implementation that follows the MCP standard and can be plugged into any AI assistant that supports MCP. The README lists a number of AI tools that already support MCP as clients, including Claude Desktop, the Cursor code editor, VS Code with Copilot, and several others. A server appearing in this list can theoretically be connected to any of those clients without writing custom glue code. The README includes a security warning worth noting. MCP servers run with the same system permissions as the application that started them, which means a malicious or poorly written server could read files, run commands, or access sensitive data. The README advises running servers in isolated environments, reviewing any server's code before installing it, and granting only the permissions each server actually needs. Official implementations from the original protocol authors are marked with a star symbol to help distinguish them from community-contributed ones. The list is a community effort and accepts contributions. The full README is longer than what was shown.
← appcypher on gitmyhub — every repo by this author, as a profile.
Verify against the repo before relying on details.