SiteFetch MCP Server

Example Use Cases

• 1

  Ingest and analyze the full content of a website so they can answer questions or generate summaries with accurate page context.

• 2

  Cached, repeatable access to site content to avoid repeated network requests and to operate faster or offline.

• 3

  Expose website content as MCP resources (sitefetch://{encodedUrl}) so that downstream models like Claude can include it in their context.

Description

This server captures entire websites using the sitefetch tool, stores the extracted content in a local cache (~/.cache/sitefetch), and provides that content to language models via the Model Context Protocol. It integrates as an MCP server and offers tools for fetching sites, listing cached sites, and clearing the cache. This enables LLM agents to reference complete site content (via sitefetch://{encodedUrl}) for analysis, summarization, or question answering without re-downloading pages repeatedly.