Fetch MCP
Official web fetch server — let AI retrieve content from any URL, including HTML to Markdown conversion for clean LLM consumption.
// Add to your client
{
"mcpServers": {
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
}Paste into your client's MCP configuration file.
// Try it now
"Fetch https://example.com and tell me what the page is about"
- ! JavaScript-heavy pages return mostly empty content — fetch only sees initial HTML
- ! Some sites block bot user agents — requests may fail or return error pages
- ! Binary content (images, PDFs) returns raw bytes which may blow context windows
API key or simple config
// When to use
- You want to read a web page or API response in conversation
- The URL returns server-rendered HTML, JSON, or plain text
- You need a lightweight, free option for single-page fetches
- You're combining with other tools and just need the page content
// When NOT to use
- The page needs JavaScript to render content — use puppeteer or browserbase
- You're crawling many pages — use firecrawl for bulk
- You need authenticated fetches with complex cookie flows — use a proper client
// Usage Scenarios
Single Page Reading
Fetch a URL and summarize or extract content.
Example prompt
"Fetch https://example.com/article and summarize the main points"
API Response Inspection
Hit a public API endpoint and inspect the response.
Example prompt
"Fetch https://api.github.com/repos/anthropic/anthropic-sdk-python and show me the star count and language"
Content Conversion
Grab a page and have AI convert it to another format.
Example prompt
"Fetch this blog post and convert it to a clean markdown file"
// About
Plain English
This server is the simplest possible way to let an AI read the web. You paste a link into the chat, the AI grabs the page, and now it can talk about what's on it. Ask Claude to summarize an article, pull the key points out of a documentation page, or look at a public API response — once Fetch is installed, those things just work. Under the hood it acts like a stripped-down browser that downloads the page and converts it to clean text the AI can read efficiently. There's no JavaScript engine and no rendering, so it's small, fast, and free. That's also the limit: pages that build themselves with JavaScript after loading (lots of modern web apps) come back nearly empty. For those you'd reach for a different tool. For most everyday tasks — reading a blog post, checking a docs page, glancing at a public JSON endpoint — Fetch is the right starting point. It's also the foundation for bigger workflows: combine it with other servers and your AI can research, extract, and summarize the open web by itself.
The official Fetch MCP server enables AI assistants to fetch and process web content. It handles HTML-to-Markdown conversion, follows redirects, and can extract specific content from pages.
A foundational building block for any AI workflow that needs to access web content, documentation, or APIs.
// Use Cases
- Read documentation pages during development
- Fetch and analyze web content for research
- Access APIs and parse responses
- Extract structured data from websites
// Works With
// Also Consider
// Related Servers
Brave Search MCP
Official Brave Search integration — give AI real-time web search access via the Brave Search API for up-to-date information.
Puppeteer MCP
Official Puppeteer browser automation server — let AI control a real browser for scraping, screenshots, and UI testing.
Playwright MCP
Official Microsoft Playwright browser automation server — AI-powered cross-browser testing and web interaction.