Agentfetch

Overview

Agentfetch is a Model Context Protocol server that provides token-budgeted web fetching for AI agents. It intelligently auto-routes requests through Jina, FireCrawl, Trafilatura, and PDF handlers, allowing any MCP-compatible AI assistant to call its tools directly during conversations and agent runs.

The server runs locally via the Python package agentfetch-mcp, installed with uvx agentfetch-mcp. It requires an API key for authentication and communicates with your machine over stdio. The latest release is v1.0.1, published under the MIT license with source code available at bch1212/agentfetch-mcp on GitHub.

Agentfetch appears in the AI & Agents category and integrates with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and others across the ecosystem. If you maintain this project, claim your listing on mcp.site to verify ownership and keep information up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "agentfetch": {
      "command": "uvx",
      "args": [
        "agentfetch-mcp"
      ]
    }
  }
}