About MCP Index

MCP Index is a structured directory of Model Context Protocol (MCP) servers and SKILL.md-based agent skills. Every entry is enriched with live GitHub metadata — stars, last-commit recency, language, license — and given a transparent heuristic quality score so you can quickly judge whether a project is production-ready or abandoned.

FAQ

What is an MCP server?

MCP (Model Context Protocol) is an open protocol that lets AI models securely use external tools and data sources. An MCP server exposes tools (APIs, databases, browsers, file systems…) to any MCP-compatible client such as Claude Desktop, Claude Code, Cursor, VS Code or Windsurf.

What is an agent skill?

An agent skill is a folder with a SKILL.md file (instructions plus optional scripts) that AI coding agents like Claude Code, Codex CLI or Gemini CLI load on demand to perform a specialized task.

How is the quality score computed?

0–100, from: log-scaled GitHub stars (max 40), maintenance activity based on last commit (max 25), license present (10), official status (10), not archived (5), meaningful description (5), topics (5). It is a heuristic signal, not an endorsement.

Where does the data come from?

Public curated lists — modelcontextprotocol/servers, punkpeye/awesome-mcp-servers, VoltAgent/awesome-agent-skills — enriched via the GitHub API. The pipeline reruns weekly; dead repos are dropped automatically. The dataset and pipeline are open source on GitHub.

How do I add or remove a listing?

Get your project added to one of the upstream curated lists, or open an issue on our GitHub repository.