Agentfetch

Please install the `agentfetch` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install Python 3.11+** — Python 3.11 or newer is required to run agentfetch-mcp. (https://www.python.org/downloads/)
- **Install agentfetch-mcp from PyPI** — Install the MCP server package using pip. This is the simplest portable install path. Run: `pip install agentfetch-mcp`

Optional prerequisites:
- Get a Jina Reader API key — Recommended for full fetching capability. Without it, only Trafilatura-based fetching works for roughly 70% of standard pages. (https://jina.ai/reader)
- Get a FireCrawl API key — Optional but recommended for JS-heavy pages such as Twitter/X, LinkedIn, and Notion.
- Set up Redis — Optional. Provide a Redis instance if you want caching; without Redis, the server runs uncached.

Canonical MCP server config (stdio transport):
- command: `python`
- args: ["-m","agentfetch.mcp.server"]
- optional environment variables:
  - `JINA_API_KEY`: Jina Reader API key. Recommended for broader web page coverage; without it, only Trafilatura works. (example: `<your-jina-api-key>`)
  - `FIRECRAWL_API_KEY`: FireCrawl API key for JS-heavy pages such as Twitter/X, LinkedIn, and Notion. (example: `<your-firecrawl-api-key>`)
  - `REDIS_URL`: Redis connection URL for caching. If omitted, caching is disabled. (example: `<your-redis-url>`)
  - `CACHE_TTL_SECONDS`: Cache TTL in seconds for fetch results. (example: `<cache-ttl-seconds>`)

Note: The README states the server speaks MCP over stdio. Although JINA_API_KEY is labeled recommended rather than required, the server can still function without it using Trafilatura only.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.