Skip to main content
MCP server CX Codex

web-search · magurelabs — Claude Code MCP server: what it does, how to install

web-search — an MCP server for Claude Code that searches the web and extracts readable text from pages.

Apache-2.0 ★ 10 updated 2026-11-16

Open on GitHub

What web-search does

The connector queries search providers and fetches text contents from web pages directly inside agent workflows. Running it requires Python 3.11 or later, along with a SearXNG instance or a Serper key, and Crawl4AI for the scrape_url tool.

The server builds on the official Python SDK for Model Context Protocol and exposes a streamable HTTP transport. Installation requires cloning the repository source code and setting up a local virtual environment.

What it helps with

  • search the web from AI agents
  • fetch readable text content from URLs
  • integrate web search tools into an MCP host

Example prompts for the agent

  • Start the search connector server via MCP_MODE=local python -m connectors.web_search
  • Query the MCP server from your client to execute web searches

How to install web-search

Install

The MCP registry has no launch command for it: see the repository README

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about web-search

How to install web-search in Claude Code?

Clone the repository with git clone https://github.com/magurelabs/magoneai-mcp, change directory to cd magoneai-mcp, create a virtual environment using python3.11 -m venv .venv && source .venv/bin/activate, and run pip install -e ..

What does web-search do?

It provides MCP tools to search the web and fetch clean text from web pages.

What is needed to run web-search?

You need Python 3.11 or later, a SearXNG instance or Serper API key, and Crawl4AI for page scraping.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

web-search by other authors comes first.

web-search keenableai web-search searches the Keenable web index and retrieves page text formatted as clean Markdown. Works without an API key for up to 1,000 requests per hour. CX Codex +1 ★ 9repo stars Keenable Web Search keenable Keenable Web Search — a remote MCP server for searching the web and retrieving page contents as markdown. Runs out of the box without API keys or accounts. CX Codex +1 2,498installs Web Search hec-ovi The Web Search MCP server for Claude Code searches multiple engines and retrieves clean Markdown without API keys for fresh data and arXiv research. CX Codex +1 ★ 12repo stars Web Search — Structured Results for AI Agents axel-belfort Web Search — Structured Results for AI Agents is an MCP server for web search in coding assistant conversations. It returns structured JSON with titles, URLs, and snippets without standard API credentials. CX Codex +1 4,334installs web-search-plus robbyczgw-cla web-search-plus searches across 11 providers with automatic query routing and extracts web page text. It switches providers instantly when one fails or lags. CX Codex +1 1,200installs tavily-search tavily-ai tavily-search — a skill for Claude Code that enables web searching and data extraction from webpages. Use it to verify facts, read current news, and gather information beyond the model's training data. CX Codex +1 39,400installs Private Web Search nickclyde Private Web Search — an MCP server that provides DuckDuckGo search and webpage fetching. Long result URLs are replaced with short ref:// tokens to save context. CX Codex +1

From the same set15

The whole set →

Project sponsors

Companies that keep this analytics open