# Team0 Living Understanding [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/Team0-ai/team0-plugins  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/team0-living-understanding

## Description
The second brain your AI agents share. It builds itself from your work and keeps itself current.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What Team0 Living Understanding does

The Team0 Living Understanding MCP server connects AI agents to Team0's shared understanding of a team's work. The repository contains public plugin packages, while the Team0 application and hosted MCP service are maintained separately.

The native agent runtime is intended to make context available across supported hosts. Before an ordinary turn, its hooks load relevant Living Understanding. After the turn, they return the completed interaction for governed learning. This gives agents a shared context source that can be kept current as work continues.

The repository also includes the source package for Team0's ChatGPT plugin. That plugin connects to Team0's hosted MCP service for on-demand reads and contributions rather than using the native before- and after-turn runtime hooks.

## How it works

The Team0 Living Understanding MCP server uses host-specific plugin adapters and a Team0 connection. The runtime package includes native adapters for Claude Code, Codex, and OpenClaw. Hermes is supported through the same runtime package with its own plugin installation and key configuration.

Each host requires its own Team0 connection and host-level approval. OpenClaw additionally requires conversation-access approval in its Gateway. Installing a package does not automatically grant trust or connection consent.

ChatGPT uses the hosted MCP service through its plugin flow. It does not automatically run the before-turn and after-turn hooks used by the native agent runtime. Claude.ai, Claude Cowork, and Town connect to the hosted service directly and do not have downloadable packages in this repository.

## Setup and configuration

Install the runtime through the host's supported mechanism:

- Claude Code: add the Team0 plugin marketplace and install `team0-agent-runtime@team0`.
- Codex: add the Team0 plugin marketplace and install `team0-agent-runtime@team0`.
- OpenClaw: install **Team0 Living Understanding** from Plugins or use `openclaw plugins install clawhub:team0-agent-runtime` in the relevant Gateway.
- Hermes: save the Team0 access key as `TEAM0_HERMES_API_KEY` under **Keys → Custom Keys**, then install the runtime from GitHub or a Git URL and enable it.

OpenClaw must have a Team0 MCP connection and conversation-access approval in its Gateway. For ChatGPT, installation and OAuth consent happen in ChatGPT; the repository folder is source code, not a ChatGPT Store listing. Start a new Hermes chat after activating the Team0 connection.

## Tools and capabilities

The Team0 Living Understanding MCP server supports these documented operations and behaviors:

- Read relevant Living Understanding before an ordinary agent turn.
- Return completed turns for governed learning.
- Perform on-demand reads through the ChatGPT plugin.
- Submit contributions through the ChatGPT plugin.
- Share Team0 context across supported agent hosts.
- Stop an agent's access or contributions from Team0's **Agents** settings.

## Limitations and notes

The repository does not contain the Team0 application or hosted MCP service. It provides plugin packages and ChatGPT plugin source. Installation commands alone do not provide trust or consent, and each host has separate approval requirements.

ChatGPT does not receive automatic before- and after-turn hooks through its plugin. Claude.ai, Claude Cowork, and Town use the hosted service without downloadable packages here. Support is available at support@team0.ai.

_Full upstream README: https://allmcps.com/mcp/team0-living-understanding/readme_