Safer Fetch

Provides web content fetching with prompt-injection safeguards, converting HTML to markdown and returning safe data for LLMs.
  • typescript

0

GitHub Stars

typescript

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "tommertom-safer_fetch_mcp": {
      "command": "uvx",
      "args": [
        "--refresh",
        "mcp-server-fetch-tom"
      ],
      "env": {
        "PYTHONIOENCODING": "utf-8"
      }
    }
  }
}

Safer Fetch MCP Server lets language models retrieve web content safely by converting HTML to markdown and wrapping content with strict prompt-injection safeguards. It helps you fetch pages for processing while minimizing the risk of prompt manipulation.

How to use

Use the fetch MCP server from your MCP client to fetch a URL and receive the content as markdown. The server applies content boundary wrapping and pattern-detection checks to reduce the chance of prompt injection. You can fetch a page, read the extracted content in chunks if needed, and review any warnings if suspicious patterns are detected.

How to install

Prerequisites: you need the MCP runtime tooling available in your environment. The server runs via the UVX runner.

uvx --refresh mcp-server-fetch-tom

Configuration and usage notes

Configure your MCP client to point at the fetch server using a stdio configuration. The server is started through a simple command that refreshes to ensure the latest version is used.

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

Notes on environment and Windows quirks

If you encounter encoding issues on Windows, you may need to set the Python IO encoding to UTF-8 in the environment for the fetch server.

{
  "mcpServers": {
    "fetch": {
      "command": "uvx",
      "args": ["--refresh", "mcp-server-fetch-tom"],
      "env": {
        "PYTHONIOENCODING": "utf-8"
      }
    }
  }
}

Available tools

fetch

Fetches a URL from the internet and extracts its contents as markdown. Supports max_length, start_index, and raw mode; automatically converts PDFs to plain text when needed.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational