About Team0 Living Understanding
Team0 Living Understanding provides shared, continuously updated context for AI agents based on team work. Its public plugins connect supported agent hosts and ChatGPT to Team0’s hosted MCP service, with runtime hooks available for several hosts.
Key features
- 01Load relevant context before agent turns
- 02Submit completed turns for governed learning
- 03Adapters for Claude Code, Codex, and OpenClaw
- 04Hermes before- and after-turn hooks
- 05ChatGPT on-demand reads and contributions
Use cases
- 01Provide agents context from shared team work
- 02Update shared understanding after agent turns
- 03Use Team0 context from ChatGPT
MCP support
The ChatGPT plugin connects to Team0’s hosted MCP service for reads and contributions. Other listed hosts connect to the same service, while native runtime hooks are provided for supported agent environments.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://api.team0.ai/v1/mcpclaude mcp add --transport http ai-team0-living-understanding https://api.team0.ai/v1/mcp{"mcpServers":{"ai-team0-living-understanding":{"type":"http","url":"https://api.team0.ai/v1/mcp"}}}Open in Cursor{"servers":{"ai-team0-living-understanding":{"type":"http","url":"https://api.team0.ai/v1/mcp"}}}Open in VS Code{"mcpServers":{"ai-team0-living-understanding":{"type":"http","url":"https://api.team0.ai/v1/mcp"}}}{"mcpServers":{"ai-team0-living-understanding":{"serverUrl":"https://api.team0.ai/v1/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMETeam0 plugins
This repository contains the public plugin packages for Team0. The Team0 application and its hosted MCP service are maintained separately.
Agent runtime: Claude Code, Codex, and OpenClaw
team0-agent-runtime/ is one shared package with native
adapters for all three hosts. Its hooks load relevant Living Understanding before
an ordinary turn and return completed turns for governed learning. Each host
needs its own Team0 connection and host-level plugin approval.
Claude Code
claude plugin marketplace add Team0-ai/team0-plugins
claude plugin install team0-agent-runtime@team0
Codex
codex plugin marketplace add Team0-ai/team0-plugins
codex plugin add team0-agent-runtime@team0
OpenClaw
In OpenClaw's Plugins, search for Team0 Living Understanding and install it on the Gateway you use. View the published plugin on ClawHub. For CLI installation in that same Gateway environment:
openclaw plugins install clawhub:team0-agent-runtime
OpenClaw also needs a Team0 MCP connection and conversation-access approval in its Gateway. See the runtime setup and host-specific instructions before enabling its hooks. None of these installation commands grants trust or connection consent automatically.
Hermes
In the active Hermes profile, open Keys → Custom Keys and save your Team0
access key as TEAM0_HERMES_API_KEY. Then open Plugins → Install from GitHub / Git URL,
enter Team0-ai/team0-plugins/team0-agent-runtime, select Enable after install,
and install. Review Hermes's security prompt yourself. Start a new chat after
activating the connection in Team0. The plugin runs before/after-turn hooks in
Hermes; no MCP connection or SOUL.md instructions are required for these hooks.
See the Hermes setup for the terminal alternative.
ChatGPT
team0-chatgpt/ is the source package for Team0's ChatGPT
plugin. It connects to Team0's hosted MCP service for on-demand reads and
contributions. Unlike the native agent runtime, ChatGPT does not run Team0
before- and after-turn hooks automatically. This GitHub folder is source code,
not a ChatGPT Store listing or a Claude/Codex marketplace entry; installation
and OAuth consent happen in ChatGPT.
Claude.ai, Claude Cowork, and Town connect to the hosted Team0 MCP service; they do not have downloadable packages in this repository.
You can stop an agent's access or contributions in Team0 under Agents. For help, contact [email protected].
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.