Connect to the javadocs.dev MCP server

This is the javadocs.dev MCP server. It gives AI coding agents tools to resolve Maven Central versions, list and read Javadoc/Scaladoc, and read library source - for any Java, Kotlin, or Scala artifact, with no local build or checkout required.

Endpoint (Streamable HTTP): https://www.javadocs.dev/mcp

Claude Code

Run in your project (add --scope user to enable it in every project):

claude mcp add --transport http javadocs https://www.javadocs.dev/mcp

Cursor

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (per-project):

{
  "mcpServers": {
    "javadocs": {
      "url": "https://www.javadocs.dev/mcp"
    }
  }
}

Kiro / Kiro CLI

Add to .kiro/settings/mcp.json (workspace) or ~/.kiro/settings/mcp.json (global):

{
  "mcpServers": {
    "javadocs": {
      "url": "https://www.javadocs.dev/mcp"
    }
  }
}

VS Code (GitHub Copilot)

Add to .vscode/mcp.json in your workspace (or run MCP: Add Server from the Command Palette):

{
  "servers": {
    "javadocs": {
      "type": "http",
      "url": "https://www.javadocs.dev/mcp"
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "javadocs": {
      "serverUrl": "https://www.javadocs.dev/mcp"
    }
  }
}

Cline / Roo Code

Edit the MCP settings JSON (cline_mcp_settings.json) or use the MCP Servers UI:

{
  "mcpServers": {
    "javadocs": {
      "type": "streamableHttp",
      "url": "https://www.javadocs.dev/mcp"
    }
  }
}

Gemini CLI

Add to ~/.gemini/settings.json (use httpUrl for Streamable HTTP):

{
  "mcpServers": {
    "javadocs": {
      "httpUrl": "https://www.javadocs.dev/mcp"
    }
  }
}

Claude Desktop, Zed, and other stdio-only clients

For clients that only speak stdio, bridge to the remote server with mcp-remote (requires Node.js):

{
  "mcpServers": {
    "javadocs": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://www.javadocs.dev/mcp"]
    }
  }
}

Prefer not to use MCP? The same operations are available as a read-only REST API (SwaggerUI), and every page returns Markdown with an Accept: text/markdown header. There's also an Agent Skill describing how to drive this service.