MCP server · Notes
threadctx MCP server
Your AI remembers past fixes, decisions and gotchas for each project, and shares them with your teammates' AI.
threadctx is a memory helper for AI coding assistants. Normally, every time you start a new chat, your AI forgets everything from last time. With threadctx connected, the useful things your team figured out along the way get written down and handed back to the AI when it matters. It is handy for developers and small teams who keep solving the same problems over and over.
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 one extra skill or one extra connection. This one connects your AI to a shared memory store for your project. So when you ask it to fix something, it can first look up what your team already learned about that part of the code.
What this MCP server does
You ask your AI to do a task, like fixing a bug or adding a feature. Before it starts, the AI quietly asks this helper whether anything similar has come up before. The helper looks in its memory file for your project and hands back the relevant notes. When the AI finishes and learns something new, it writes that note back into the memory. Next time, you or a teammate gets that knowledge without anyone having to explain it again.
Click to zoomWhat you can do with it
- Remember fixes for tricky bugs so you do not solve them twice
- Store architectural decisions your team made and why
- Share those notes with teammates who use the same project
- Look up past gotchas before your AI starts risky work
- Turn recent git commits into memory notes automatically
- See everything stored on your machine with a simple list command
- Set up a whole project for your team with one command
Try asking your AI
- “Before you touch the login code, check team memory for anything we learned about it”
- “Write down that we switched from Redis to Postgres for sessions, and why”
- “What did we already try for the slow dashboard query?”
- “Capture the decisions from the last few commits into memory”
What it gives back to you
You get short notes back in the chat, usually a few lines each, with the fix or decision and a small footer saying it came from threadctx. When the AI writes a new memory, you see a short confirmation. The list command shows you everything stored on your machine, per project or across all projects. In cloud mode, the same notes come from everyone on your team.
Before you start
What you need
- Node.js 18 or newer on your computer
- An MCP client like Claude Code or Cursor
- For shared team memory: a paid threadctx Team plan and an API key from threadctx.dev
- For the automatic git capture feature: your own Anthropic or OpenAI key
Good to know
In cloud mode your notes are shared with everyone on the team, so do not store secrets or anything private in them.
Install it with your AI
Add threadctx 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 threadctx 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.
Who it's for
Developers and small engineering teams who use an AI coding assistant and are tired of re-explaining the same project history.
