> ## Documentation Index
> Fetch the complete documentation index at: https://docs.oncortex.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Team patterns

> A few ways teams put Cortex to work

Here are some common shapes that work well. Treat them as starting points.

## A team knowledge base

Create one shared brain, "Team Handbook" or similar, using the [Company template](/brains/templates). Add the team as members: most people as Write, a couple as Admin. Everyone files what they learn into it, or has their agent do so, and everyone can recall from it.

Because filing follows [skills](/filing/skills), the handbook stays consistent no matter who contributes. Start it on Suggest so the team stays in control of what lands, and move it to Automatic once you trust it.

## A client or project brain

Create a shared brain per client or project. Point relevant [meeting transcripts](/capture/meeting-transcripts) straight at it, add the people working on that account as members, and keep everything about that relationship in one place: contacts, decisions, commitments, history. When the project ends, the brain is a complete, searchable record.

## Meetings into a shared brain

One person points their meeting tool at a shared project brain (via the [transcript format](/capture/meeting-transcripts)), and the whole team's view of that project's meetings builds up in one place. Others can promote their own relevant meeting notes in from their own brains as needed.

## Private first, shared deliberately

The pattern underneath all of these: people capture freely into their own brains, and [promote](/brains/promoting) the parts that belong to the team. Nobody has to self-censor while capturing, and the shared brains only ever contain what someone chose to put there.