Context7 MCP Server

✓ verified

Up-to-date, version-specific docs for libraries and frameworks.

Don't have ToolPiper? We'll take you to the download.

Open repository →

Publisher

Upstash

License

MIT

Transport

Local subprocess

Tools

2

What this server does

Context7 fixes the stalest part of coding with an LLM: library docs. It pulls current, version-specific documentation and code examples for thousands of packages and injects them into the model's context, so the assistant stops hallucinating APIs that shipped after its training cutoff. Ask it to resolve a library, then fetch the docs for the version you actually use. It works without a key on a rate-limited free tier; add a Context7 API key for higher limits.

Tools included

  • resolve_library_id

    Resolve a library name to a Context7 ID.

  • get_library_docs

    Fetch current, version-specific docs for a library.

The authoritative list comes from the server's tools/list at runtime. This catalog entry may lag the upstream until the next ToolPiper release.

Install in ToolPiper

ToolPiper handles the JSON configuration. Click Install, fill in any required tokens, and the server is ready in your chat.

Don't have ToolPiper? We'll take you to the download.

Install in another client

Install in another MCP client (Claude Desktop, Cursor, Continue, Cline)

Add this to ~/Library/Application Support/Claude/claude_desktop_config.json.

{
  "mcpServers": {
    "context7": {
      "command": "npx",
      "args": [
        "-y",
        "@upstash/context7-mcp"
      ]
    }
  }
}

Example prompts

  • "Get the current docs for the Next.js App Router and show me how to set up a route handler."

  • "How do I configure retries in the latest version of the AWS SDK for JavaScript?"

  • "Pull the Tailwind v4 config reference and explain the @theme directive."

Security notes

Works keyless on a rate-limited free tier, so no credential is required to start. An optional CONTEXT7_API_KEY raises limits; if you add one, treat it as a credential. The server only reads public documentation, so the privacy posture is low-risk; the content it returns is third-party docs, not your code.

Related tools