Repo Memory MCP server

Repo Memory is a small helper that gives your AI a notebook about your code project. When your AI figures out something useful, like where a piece of code lives or why a choice was made, it writes that down in a folder inside your project. The next time you or a teammate opens the project with any AI tool, it reads those notes first instead of starting from scratch.

What this MCP server does

You ask your AI something about your project, like where the login code lives. The AI uses this helper to read the notes folder inside your project. If the answer is already there, it tells you right away without searching again. If the AI discovers something new while working, it can add a note, a decision or a warning to that folder. When you commit and push your project with git, those notes travel to your teammates too.

Click to zoom

What you can do with it

• Read everything your AI already learned about a project
• Save a verified fact, like where a file or function lives
• Record why a technical choice was made
• Write down a warning about something that wasted time
• List only the facts about one area, like login or payments
• Share all of this with teammates by committing the notes folder

Try asking your AI

• “Before we start, check the repo memory for anything about the login flow.”
• “Save a fact that the payment code lives in src/billing/charge.py around lines 30 to 90.”
• “Write down a decision that we picked this library because it supports async.”
• “List all the gotchas we have recorded so far.”

What it gives back to you

You get back plain answers in the chat, like a short summary of what was already known or a list of saved facts. When your AI saves something, it confirms that the note was added. The notes themselves live as simple text files inside your project folder, so you can open and read them yourself anytime.

Before you start

Who it's for

Developers and small teams who use AI coding tools on the same project and are tired of explaining the same things over and over.