mcp-fetch by LangGPT - MCP Server

MCP Fetch

A Model Context Protocol server that provides web content fetching capabilities with robots.txt checking removed. This server enables LLMs to retrieve and process content from web pages, converting HTML to markdown for easier consumption.

This is a modified version of the original mcp-server-fetch that removes all robots.txt checking, allowing unrestricted access to web content.

[!CAUTION] This server can access local/internal IP addresses and may represent a security risk. Exercise caution when using this MCP server to ensure this does not expose any sensitive data. Additionally, this version ignores robots.txt restrictions which may violate some websites' access policies.

The fetch tool will truncate the response, but by using the start_index argument, you can specify where to start the content extraction. This lets models read a webpage in chunks, until they find the information they need.

Available Tools

fetch - Fetches a URL from the internet and extracts its contents as markdown.
- url (string, required): URL to fetch
- max_length (integer, optional): Maximum number of characters to return (default: 5000)
- start_index (integer, optional): Start content from this character index (default: 0)
- raw (boolean, optional): Get raw content without markdown conversion (default: false)

Available Prompts

fetch
- Fetch a URL and extract its contents as markdown
- Arguments:
  - url (string, required): URL to fetch

Installation and Usage

Local Development Setup

Clone or download the source code:

git clone https://github.com/LangGPT/mcp-fetch.git
cd mcp-fetch

Install dependencies using uv:
```
uv sync
```
Test the server:
```
uv run python -m mcp_fetch --help
```

Using with Claude Desktop (Local Source)

Create Claude Desktop configuration:

{
  "mcpServers": {
    "mcp-fetch": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/your/mcp-fetch",
        "python",
        "-m",
        "mcp_fetch"
      ]
    }
  }
}

Add configuration to Claude Desktop:
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
- Windows: %APPDATA%/Claude/claude_desktop_config.json
- Linux: ~/.config/Claude/claude_desktop_config.json
Restart Claude Desktop to load the new server.

Using with VS Code (Local Source)

Add to your VS Code settings or .vscode/mcp.json:

{
  "mcp": {
    "servers": {
      "mcp-fetch": {
        "command": "uv",
        "args": [
          "run",
          "--directory",
          "/path/to/your/mcp-fetch",
          "python",
          "-m",
          "mcp_fetch"
        ]
      }
    }
  }
}

Installation via Package Manager

Using uv (recommended)

When using uv no specific installation is needed. We will use uvx to directly run mcp-fetch:

uvx mcp-fetch

Using pip

pip install mcp-fetch

After installation, run it as:

python -m mcp_fetch

Package Manager Configuration

Claude Desktop with uvx

{
  "mcpServers": {
    "mcp-fetch": {
      "command": "uvx",
      "args": ["mcp-fetch"]
    }
  }
}

VS Code with uvx

{
  "mcp": {
    "servers": {
      "mcp-fetch": {
        "command": "uvx",
        "args": ["mcp-fetch"]
      }
    }
  }
}

Development

Setting up Development Environment

Install development dependencies:
```
uv sync --dev
```
Run linting and type checking:
```
uv run ruff check
uv run pyright
```
Build the package:
```
uv build
```

Testing

Test the server locally:

uv run python -m mcp_fetch

Use the MCP inspector for debugging:

npx @modelcontextprotocol/inspector uv run python -m mcp_fetch

Making Changes

Edit the source code in src/mcp_fetch/
Test your changes with uv run python -m mcp_fetch
Update version in pyproject.toml if needed
Run tests and linting

Publishing

Publishing to PyPI

Build the package:
```
uv build
```

Publish to PyPI:

uv publish

Or using twine:

pip install twine
twine upload dist/*

Publishing to GitHub

Initialize git repository (if not already done):
```
git init
git branch -m main
```

Add and commit files:

git add .
git commit -m "Initial commit: MCP Web Fetch server without robots.txt checking"

Create GitHub repository and push:

# Create repository on GitHub first, then:
git remote add origin https://github.com/LangGPT/mcp-fetch.git
git push -u origin main

Create a release on GitHub:
- Go to your repository on GitHub
- Click "Releases" → "Create a new release"
- Tag version: v0.6.3
- Release title: v0.6.3 - MCP Fetch
- Describe your changes
- Publish release

Building Docker Image

docker build -t mcp-fetch .
docker tag mcp-fetch LangGPT/mcp-fetch:latest
docker push LangGPT/mcp-fetch:latest

Customization

robots.txt

This version has robots.txt checking completely removed. All web requests will proceed regardless of robots.txt restrictions.

User-agent

By default, depending on if the request came from the model (via a tool), or was user initiated (via a prompt), the server will use either the user-agent:

ModelContextProtocol/1.0 (Autonomous; +https://github.com/modelcontextprotocol/servers)

or:

ModelContextProtocol/1.0 (User-Specified; +https://github.com/modelcontextprotocol/servers)

This can be customized by adding the argument --user-agent=YourUserAgent to the args list in the configuration.

Proxy

The server can be configured to use a proxy by using the --proxy-url argument.

Debugging

You can use the MCP inspector to debug the server:

For local development:

npx @modelcontextprotocol/inspector uv run python -m mcp_fetch

For uvx installations:

npx @modelcontextprotocol/inspector uvx mcp-fetch

Contributing

We encourage contributions to help expand and improve mcp-fetch. Whether you want to add new tools, enhance existing functionality, or improve documentation, your input is valuable.

License

mcp-fetch is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.