# The Team Second Brain

The second brain is a personal system. Tiago Forte's [Building a Second Brain](/blog/what-is-a-second-brain) taught a generation of knowledge workers to capture what they learn in one trusted place. It works, for one person.

Teams never got the equivalent. Your company's real knowledge, the decisions, the runbooks, the reasons things are the way they are, lives scattered across a few people's heads, a wiki nobody opens, and a hundred Slack threads. Everyone has their own notes. Nobody has the team's.

A team second brain is that missing piece: one shared place your whole team writes down what it knows, and, in the AI era, one that every teammate's Claude and ChatGPT can read.

## Why now, and not five years ago

The personal second brain always had a weak spot. You still had to go find the note. Capture was easy; retrieval at the moment you needed it was the hard part.

AI closes that gap, if the notes are somewhere it can reach. Connect a knowledge base to Claude or ChatGPT with a [built-in MCP server](/docs/what-is-mcp) and your assistant can pull the right note into the conversation itself, instead of you remembering it exists. That is what makes a second brain worth building for a team now: the payoff isn't just an organized archive, it's that every teammate's AI can answer from it. This is the [AI-native second brain](/blog/the-ai-native-second-brain), pointed at a team instead of a person.

## What goes in a team second brain

The knowledge that currently lives in people, not documents.

- **Decisions.** Why you chose what you chose, and what you rejected, so nobody relitigates it.
- **Runbooks and process.** How you deploy, release, and respond when things break.
- **Context.** The customers, the history, the conventions a new person takes months to absorb.
- **Ownership.** Who owns what, and who to ask.

Written once, in a shared space, where every teammate's assistant can read it and the team keeps it current together.

## But our team wiki is a graveyard

Most are. A team wiki dies because writing to it never pays the writer back, so nobody bothers, so it goes stale, so nobody reads it.

A team second brain breaks that loop because the AI reads it. When a teammate asks their assistant a question, it can answer from the notes, which means keeping a note current finally pays off the moment someone asks. The archive stops being a chore and becomes something the team's assistants actually consult. Same notes serve Claude and ChatGPT, so it doesn't matter which one each person uses.

## Building one

1. Create a team and folders for decisions, runbooks, and context. The [set up your team](/docs/set-up-a-team) guide walks through it.
2. Write down the things people ask each other most. That's the first version.
3. Each teammate connects their own Claude or ChatGPT, and a team is free for its first 25 team notes. Now the team's brain is one place every assistant can read.

The [full how-to is here](/blog/team-knowledge-base-for-claude-and-chatgpt), and [the company brain](/company-brain) page shows the shape it takes across a whole company.

## The point

You already built a second brain for yourself. Your team is running without one, on memory and Slack search and the two people who have been here longest.

Give it one every assistant can read. See [Hjarni for teams](/for-teams).

Write once. The whole team remembers.