Skip to main content
Glama
mcp-safe-fetch

mcp-safe-fetch

by PortfolioKB

mcp-safe-fetch

tests

An injection-aware content fetcher, exposed over the Model Context Protocol.

Agents that read the open web read content nobody on your team wrote. A page or a post can carry a fake closing tag followed by "ignore previous instructions and ...". This server fetches a URL, strips it to text, runs defense-in-depth sanitization, caps the size, and caches the result. It is small on purpose. It is the reference implementation of a set of production hardening field notes, not a framework.

Why it exists

Most MCP servers are written for a demo: one caller, one happy path, no untrusted input, no cost ceiling. In production the failures cluster in three places, and none of them show up on day one:

  • Untrusted content. A single tag stripper feels safe and is not. Unicode variants and unclosed tags walk straight through one regex.

  • Cost. An unbounded response body or an oversized context is money spent on noise.

  • Concurrency. Two tools touching one SQLite file throw database is locked.

This server answers all three in code you can read in five minutes.

Related MCP server: WebSurfer MCP

What it does

fetch_clean(url, max_chars=2500) returns sanitized, size-capped text plus an audit of what was done. The defense is order, not cleverness:

  1. strip the injection wrapper by its literal name, first

  2. normalize unicode (NFKC) so homoglyph tags cannot hide

  3. drop script and style bodies, then the generic tag strip

  4. a second net for common instruction-override phrases

It caps input size with a [truncated] marker (if the model has to ignore most of the input, you are paying for nothing), clamps the per-call cap at the entry to a hard ceiling, and caches results in SQLite opened with WAL and busy_timeout so overlapping callers wait instead of crashing.

Install

pip install -e .

Run

As a standalone MCP server (stdio):

mcp-safe-fetch

Register it with an MCP client (for example, Claude Code) by pointing the client at the mcp-safe-fetch command. The server exposes one tool, fetch_clean.

Test

pip install -e ".[dev]"
pytest

The field notes behind it

The reasoning, with the production incidents that motivated each defense, is written up here: a short essay on MCP hardening (concurrency, prompt injection, cost) and what breaks after day 30. The sanitizer in src/mcp_safe_fetch/sanitize.py is the exact function from that write-up.

License

MIT.

Available Tools

1 tool
fetch_cleanB

Fetch a URL and return injection-sanitized, size-capped text.

Args: url: an http/https URL to fetch. max_chars: per-fetch character cap (clamped to a hard ceiling).

Returns a dict with the cleaned content and an audit of what was done.

Inputs: url (required), max_chars.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 1 tool

    Disambiguation5/5

    Only one tool exists, so there is no ambiguity at all. An agent cannot misselect between tools.

    Naming Consistency5/5

    With a single tool, naming is trivially consistent. The name 'fetch_clean' follows a clear verb_noun pattern.

    Tool Count3/5

    One tool feels thin for a server dedicated to fetching content. While the tool is well-defined, the server would benefit from additional tools (e.g., batch fetch, status check) to justify its scope.

    Completeness4/5

    The tool covers the core fetch-and-sanitize operation completely for a single URL. Minor gaps exist (e.g., no way to fetch multiple URLs or configure sanitization levels), but the basic use case is fully served.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Securely fetches web content, extracts links and metadata, and downloads files through a sandboxed MCP server without JavaScript execution. Includes prompt-injection detection and comprehensive HTML sanitization for safe web data retrieval.
      2
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Efficiently loads, caches, and delivers project file context for AI agents using the Model Context Protocol, reducing multiple file reads to a single call with LRU caching and security validation.
      3
      24 npm
      MIT