MCP Catalogs
Home

webfetch-mcp

by manooll·26·Score 41

An MCP server providing real-time web access with tunable search and clean content extraction for local AI models.

web-scrapingsearchai-llm
8
Forks
3
Open issues
7 mo ago
Last commit
2d ago
Indexed

Overview

WebFetch.MCP is a self-hosted MCP server that bridges the gap between local LLMs and the live web. It offers both web search functionality and URL content fetching capabilities without requiring API keys. The server uses SearxNG as a metasearch engine and Mozilla Readability for content extraction, with support for 70+ search engines including specialized ones like Google Scholar, arXiv, and GitHub. The implementation includes smart rate limiting (12 calls per 5 minutes) and configurable options for source weighting, content filtering, and domain restrictions.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

you:Research assistants accessing the latest academic papers and technical documentation
you:Local AI models staying current with news and real-time information
you:Content extraction for analysis while filtering out ads and navigation elements
you:Scientific research across multiple specialized databases with targeted queries
you:Does WebFetch.MCP require API keys?
you:What rate limits are imposed on the server?

When to choose this

Choose WebFetch.MCP when you need privacy-focused web access for local AI models without API keys, requiring fine-tuned search capabilities across technical and academic sources.

When NOT to choose this

Don't choose this if you need JavaScript-heavy content extraction (requires JSDOM) or require real-time content that needs immediate processing, as it has rate limiting (12 calls per 5 minutes).

Tools this server exposes

2 tools extracted from the README
  • web_search

    Search the web with 70+ configurable engines for fresh, real-time data

  • web_fetch

    Extract clean content from a specific URL using Mozilla Readability

Comparable tools

mrkrsl-web-searchmcp-server-fetch-pythoncrawl4ai

Installation

Installation

Prerequisites

  • Node.js 18+
  • Docker & Docker Compose
  • SearxNG instance

Steps

  1. Install SearxNG:
git clone https://github.com/searxng/searxng-docker.git
cd searxng-docker
sed -i "s|ultrasecretkey|$(openssl rand -hex 32)|g" searxng/settings.yml
docker compose up -d
  1. Install WebFetch.MCP:
git clone https://github.com/manull/webfetch-mcp.git
cd webfetch-mcp
npm install
node server.mjs
  1. Configure in LM Studio or Claude Desktop:
{
  "mcpServers": {
    "webfetch": {
      "command": "node",
      "args": ["/full/path/to/webfetch-mcp/server.mjs"],
      "env": {
        "SEARXNG_BASE": "http://localhost:8080"
      }
    }
  }
}

FAQ

Does WebFetch.MCP require API keys?
No, it uses SearxNG as a metasearch engine and doesn't require any API keys for search functionality.
What rate limits are imposed on the server?
WebFetch.MCP implements a smart rate limit of 12 calls per 5 minutes with 8 calls per 30-second burst to prevent spam while allowing reasonable research workflows.

Compare webfetch-mcp with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.