MCP server · Notes
Tribal memory MCP server
Your AI can save and look up the reasoning behind your team's decisions, not just the code.

Tribal is a memory helper you run yourself. It keeps the unwritten knowledge of a project, like why a decision was made or which trick fixed a nasty bug. It is handy for engineering teams who keep losing the same context in chat threads and tickets.
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. Tribal is that helper, and it connects your AI to a private memory store for your project. Once it is wired up, your AI can save notes into that store and ask it questions when you need the old context back.
What this MCP server does
You tell your AI something worth remembering, like why you chose one approach over another. The AI sends that note to Tribal, which saves it and links it to related notes. Later, when you ask about a decision or a bug, the AI asks Tribal, and Tribal searches its store and finds the most relevant items. You get a short answer or a list of related notes right in the chat. Over time, this builds a small map of your project's reasoning that your whole team can query.
Click to zoomWhat you can do with it
- Save a decision and the reason behind it
- Ask why a certain choice was made
- Find notes that support or contradict each other
- Look up a fix that worked for a past bug
- Browse related items around a topic
- Rate how useful a search result was
- Keep project knowledge in one place your team can query
Try asking your AI
- “Save this: we picked Postgres over MySQL because of the vector extension.”
- “Why did we drop the old caching layer last quarter?”
- “Find anything we wrote about the login timeout bug.”
- “Show me notes that contradict the decision to use a monorepo.”
What it gives back to you
You get answers in plain chat: a short summary, a list of related notes, or the exact item you asked for. When you save something, Tribal confirms it stored the note and shows what it linked it to. When you search, you see the most relevant items with a little context around each one. Everything stays in your own database, not on someone else's servers.
Before you start
What you need
- A Postgres 14 or newer database with the pgvector extension
- A way to make embeddings, either a local Ollama install or an API key from OpenAI or Anthropic
- An AI app that supports MCP servers, like Claude Desktop or another agent harness
- A git repository if you want to scope the memory to one project
Good to know
Tribal stores your team's notes in a database you run, so keep that database and its credentials safe and back it up if the knowledge matters.
Install it with your AI
Add Tribal memory 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 Tribal memory 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
Engineering teams and developers who want their AI to remember the reasoning behind their code, not just the code itself.





