Complete AI Training

MCP server · Notes

Knowl MCP server

by dat999zx

Give your coding AI a memory that updates itself, so it stops repeating facts you already changed.

Flow diagram: you ask your AI “What database does this project use right now?”, on your own computer the Knowl MCP server works with your own computer, and you get back A short list of matching facts.

Knowl is a memory helper for coding assistants like Claude Code, Cursor and Codex. It remembers facts about your project and retires old ones when they change, so your AI reads the current answer instead of a stale one. It is handy if you keep a CLAUDE.md file that only ever grows and starts contradicting itself.

What is an MCP server? The 30-second version

On its own, your AI can only chat with you. An MCP server is a small helper program that gives your AI a new skill or a connection to something else. Knowl is that helper for your project's memory: it runs on your computer, keeps a store of facts about your code, and lets your AI look things up or save new findings when you ask. You do not need an API key or an account.

What this MCP server does

You ask your AI a question about your project, like which database you use now. The AI asks Knowl first, before reading your files. Knowl looks through its stored facts and hands back the current answer, not the one you replaced months ago. When your AI learns something new, it saves it to Knowl, and if that fact replaces an old one, the old one is retired instead of sitting there and confusing everyone. You can also browse everything yourself in a local viewer.

Flow diagram: you ask your AI “What database does this project use right now?”, on your own computer the Knowl MCP server works with your own computer, and you get back A short list of matching facts. Click to zoom

What you can do with it

  • Ask what your project currently uses without reading every file
  • Save a decision your team made so future sessions remember it
  • Correct an outdated fact so your AI stops repeating it
  • Look up the history of a fact that changed over time
  • Keep notes about goals, constraints and architecture in one place
  • Share one memory across parallel worktrees without extra setup
  • See everything stored in a local viewer

Try asking your AI

  • “What database does this project use right now?”
  • “Remember that we moved from PostgreSQL to SQLite last week.”
  • “What decisions have we made about authentication?”
  • “Show me the history of the deployment target fact.”

What it gives back to you

Knowl gives your AI a short list of matching facts, each with its category, status and where it came from. Your AI then reads that and answers you in normal words. When it saves something, you get a quiet confirmation that the fact was stored or that an older one was retired. If Knowl is unsure whether a new fact replaces an old one, it keeps both and tells you the command to settle it.

Before you start

What you need

  • Node.js 22 or later
  • A coding assistant that supports MCP, like Claude Code, Cursor or Codex
  • About 53 MB of free space if you want the local search model (optional, keyword search works without it)

Good to know

Knowl stores facts about your project on your computer, so anything you save could be read by anyone with access to that folder, and it can retire old facts when it thinks they are replaced.

Install it with your AI

Add Knowl 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 Knowl 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 4,066 MCP servers here for you.

Sign in Become a member

Who it's for

Developers and technical teams who use AI coding assistants daily and are tired of repeating the same project facts in every new session.