Fetch MCP Server
by modelcontextprotocol
The Fetch MCP Server provides web content fetching capabilities. It enables LLMs to retrieve and process content from web pages, converting HTML to markdown for easier consumption.
Last updated: N/A
What is Fetch MCP Server?
The Fetch MCP Server is a tool that allows Large Language Models (LLMs) to fetch and process content from web pages. It converts HTML content into markdown format, making it easier for LLMs to understand and utilize the information.
How to use Fetch MCP Server?
The server provides a fetch tool that accepts a URL as input. You can configure the server to use different installation methods like uv, PIP, or Docker. The server can be customized to ignore robots.txt, use a custom user-agent, or utilize a proxy.
Key features of Fetch MCP Server
Fetches content from URLs
Converts HTML to markdown
Supports content truncation with start index
Customizable user-agent and proxy settings
Can be configured to respect or ignore robots.txt
Use cases of Fetch MCP Server
Enabling LLMs to access real-time information from the web
Providing LLMs with context from web pages for improved responses
Automating data extraction from websites
Building LLM-powered web scraping tools
FAQ from Fetch MCP Server
How do I install the Fetch MCP Server?
How do I install the Fetch MCP Server?
You can install it using uv (recommended), PIP, or Docker. Refer to the installation instructions in the README for detailed steps.
How can I configure the server to use a proxy?
How can I configure the server to use a proxy?
You can use the --proxy-url argument when running the server. Add this argument to the args list in your configuration.
Can I customize the user-agent?
Can I customize the user-agent?
Yes, you can customize the user-agent by adding the argument --user-agent=YourUserAgent to the args list in the configuration.
Does the server obey robots.txt?
Does the server obey robots.txt?
By default, the server obeys robots.txt for requests originating from models (via tools) but not for user-initiated requests. You can disable this by adding the argument --ignore-robots-txt to the args list in the configuration.
How do I debug the server?
How do I debug the server?
You can use the MCP inspector to debug the server. Refer to the debugging section in the README for instructions specific to uvx installations.