MCP server · Research
searxNcrawl MCP server
Lets your AI search the web through your own SearXNG or You.com and turn pages into clean Markdown.
searxNcrawl is a small helper program that gives your AI two new skills: searching the web and reading web pages. It runs on your own computer, and with a SearXNG search instance it does not need any paid account. It is handy if you want your AI to look things up and read full pages for you instead of guessing.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you using what it already learned. An MCP server is a little helper program that adds a new skill or a connection to an outside service. This one connects your AI to a search engine and a web browser, so it can actually look things up and read pages. You ask in normal words, and the helper does the looking for you.
What this MCP server does
You ask your AI something that needs fresh information from the web. The AI uses this helper to run a search, either through your own SearXNG search page or through You.com, and gets back a merged list of results. Then the AI can pick a result and ask the helper to open that page and turn it into clean text. The helper uses a real browser behind the scenes, so pages that load with JavaScript still work. You get the answer in the chat, based on the actual page content.
What you can do with it
- Search the web and get a merged list of results with their sources
- Turn one or many web pages into clean Markdown text
- Crawl a whole site, like a documentation site, up to a depth and page limit
- Filter search results by language and time range
- Strip repeated blocks and links to keep the text short
- Crawl pages that sit behind a login, using a saved session
Try asking your AI
- “Search for the Crawl4AI quick start and summarise how to run a first crawl”
- “Find recent news about electric cars from the last week and give me the main points”
- “Crawl the official Model Context Protocol intro page and explain it in simple words”
- “Crawl the documentation site for this tool and list the main sections”
What it gives back to you
For a search, you get a merged list of results, each showing its title, link and which provider it came from. For a crawl, you get the page content as clean Markdown or as structured data, with metadata and references. In the chat this usually shows up as a readable summary or the full text, ready to read or copy. If you crawl a whole site, you get the pages it followed, up to the limits you set.
Before you start
What you need
- uv, a small tool that runs Python programs (the README shows how to get it)
- For search: your own SearXNG instance with JSON output turned on, or a You.com API key
- For crawling: nothing extra, but the first crawl downloads a browser of about 550 MB
- An MCP client like Claude Desktop, Claude Code, Cursor, Codex or VS Code
Good to know
Crawling pages behind a login uses a saved browser session, so keep that session file private and only crawl sites you are allowed to read.
Install it with your AI
Add searxNcrawl MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check searxNcrawl MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,609 MCP servers here for you.
Who it's for
People who want their AI to search and read the web on their own machine, like researchers, writers, analysts and anyone tired of paying per search.
