RAG Documentation MCP server

This is a small helper that lets your AI look things up in documentation you have collected, instead of guessing. You add pages or docs to it once, and later your AI can search them and quote the parts that matter. It is handy if you often ask your AI about a product, a tool, or an internal guide.

What this MCP server does

You ask your AI a question in normal words. The AI sends your question to this helper. The helper searches the documentation you have stored, using meaning rather than just exact words, and finds the closest matches. It sends those excerpts back to the AI, which then writes an answer using them. You can also use it to add new pages, check what is waiting to be processed, and remove sources you no longer want.

Click to zoom

What you can do with it

• Search your stored documentation with a plain question
• See which documentation sources are already indexed
• Pull all the links off a web page and queue them for indexing
• Check which pages are still waiting to be processed
• Run the queue so waiting pages get indexed
• Clear the queue if you added the wrong things
• Remove documentation sources you no longer need

Try asking your AI

• “Search my docs for how to reset a user password”
• “What documentation sources do you have right now?”
• “Take this page and add its links to the queue: https://example.com/docs”
• “What is still waiting in the processing queue?”

What it gives back to you

You get back short excerpts from your documentation, ranked by how well they match your question, along with where they came from. For the listing and queue tools, you get simple lists, like source titles with their links and last update times, or the URLs still waiting to be processed. The AI then turns those into a normal answer in the chat. When you remove or clear something, it just tells you it is done.

Before you start

Who it's for

People who keep asking their AI about a specific product, tool, or guide and want answers grounded in real documentation.