Multi Fetch MCP Server

@alexyangjie

リポジトリ

基本情報

カテゴリ

ブラウザ自動化

ライセンス

MIT license

トランスポート

stdio

作者

alexyangjie

Multi Fetch MCP Server について

Extending mcp-server-fetch with parallel browsing

Multi Fetch MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "fetch": {
      "command": "uvx",
      "args": [
        "mcp-server-multi-fetch"
      ],
      "env": {
        "FIRECRAWL_API_KEY": "",
        "FIRECRAWL_API_URL": ""
      }
    }
  }
}

Multi Fetch MCP Server のツール

Fetches a URL from the internet and extracts its contents as markdown.

Fetches multiple URLs concurrently and returns an array of results. Input is an array of objects, each with:

Searches the web using the Firecrawl search API and scrapes results in markdown and link formats by default.

概要

What is Multi Fetch MCP Server?

Multi Fetch MCP Server is a fork of the Fetch MCP Server that replaces direct HTTP fetching with the Firecrawl Python SDK. It enables LLMs to fetch web content, scrape multiple URLs concurrently, and search the web, converting HTML to markdown for easier consumption.

How to use Multi Fetch MCP Server?

Install via pip install mcp-server-multi-fetch firecrawl-py and set the FIRECRAWL_API_KEY environment variable. Then configure Claude.app (or any MCP client) using uvx mcp-server-multi-fetch. Use the fetch, fetch_multi, and search tools to retrieve and process web content.

Key features of Multi Fetch MCP Server

Based on the Fetch MCP Server, uses the Firecrawl SDK
Fetch single or multiple URLs concurrently
Web search via the Firecrawl search API
Content truncation with start_index for chunked reading
Customizable user-agent, proxy, and robots.txt behavior
Prompts for fetch and search operations

Use cases of Multi Fetch MCP Server

Reading large web pages in chunks to find specific information
Concurrently fetching multiple sources for research or comparison
Searching the web and scraping results into markdown
Integrating web content into AI‑powered workflows

よくある質問

What is the difference from the original Fetch MCP Server?

This fork replaces direct HTTP fetching with the Firecrawl Python SDK, requiring a Firecrawl API key (`FIRECRAWL_API_KEY`) instead of using plain HTTP requests.

What dependencies are required?

Requires Python with the `mcp-server-multi-fetch` and `firecrawl-py` packages. Optionally, Node.js can be installed for a more robust HTML simplification step.

How do I set up the Firecrawl API key?

Set the `FIRECRAWL_API_KEY` environment variable to your Firecrawl API key. Optionally, override the API endpoint with `FIRECRAWL_API_URL`.

Can I configure a proxy?

Yes, use the `--proxy-url` argument with HTTP, HTTPS, or SOCKS5 proxy URLs (e.g., `--proxy-url http://192.168.1.1:8080`).

How does robots.txt handling work?

By default, the server obeys a website’s robots.txt for model‑initiated requests (tools) but not for user‑initiated requests (prompts). This can be disabled with the `--ignore-robots-txt` flag.

「ブラウザ自動化」の他のコンテンツ

QuantumProxies — Residential, Mobile & Datacenter Proxy MCP Server

quantumproxies

Give your AI agent its own proxies. QuantumProxies exposes residential, mobile, datacenter, ISP and IPv6 endpoints as MCP tools: list your plans, generate ready-to-use proxy strings with credentials, target by country, s

Loadster

loadster

Loadster is a load testing, stress testing, and synthetic monitoring platform. Run Playwright, Browser Bots, and Protocol Bots to test and monitor your websites, web apps, and APIs. Loadster MCP gives your agents the abi

Parse

Parse.bot

Parse allows your agent to access thousands of APIs instantly. You can use Parse to build an API for any action on any website and edit any previously existing API in our catalogue.