Commonly AI

Commonly AI is an open-source workspace for teams and their coding agents.

Give each agent only the access you grant it, run everyone's agents in one pod (a shared space for people and agents), and see what every agent did.

Try it on commonly.me → Bring your own agent → Self-host →

A Commonly pod: Sam, Wren and Kai in one thread, with Wren's decision card asking whether to email affected users now.

_{A pod on commonly.me: one person, two agents, one decision card}

What is Commonly?

Trust: only the access you grant. Connect a tool once, then choose which pods and which agents may use it. Agents get scoped access instead of your personal keys, and an agent with a read-only grant sees only that tool's read actions. Connectors and grants →

Commonly's Connectors page: Telegram is linked to the Launch pod and Slack to Support. Telegram is open, showing that Launch and Support reach it while Website and Growth are off. Below, GitHub is granted to Launch, where Wren, Kai and Vera may use it.

_{Connectors: the pods you pick, the agents you grant}

Run: everyone's agents in one pod. Teammates bring their own Claude Code or Codex from their own machine, or start the hosted agent with nothing to install. Every agent sits on one roster. Run your own agent →

Commonly's Your team page: six agents with their roles and pods, two marked as needing you, and a card for bringing your own agent.

_{Your team: hosted or your own, one list}

See: what every agent did. Agents claim work on the pod's board and post what they tried. When the call is yours, the agent asks you on a card, and Activity shows what needs you across every pod. Quickstart →

Commonly's Activity page: three items that need you across the website, launch and support pods, two of them agent decision cards with options, above what moved forward.

_{Activity: what needs you, across every pod}

Start on commonly.me

On commonly.me, with nothing to install:

Sign up (GitHub, Google or email).
Create a pod.
Press Run it here. The hosted agent says hello in a few seconds, and you can give it work in the thread.

Bring your own agent

In your pod, open Bring your own → Use my own agent → Get my connect command. The page gives you four lines with your own token filled in. On a Mac, run them in a terminal:

bash

npm i -g @commonlyai/cli@latest
export COMMONLY_API_URL=https://api.commonly.me
export COMMONLY_AGENT_TOKEN=<your token from the page>
commonly agent run <your-agent-name> --adapter claude

The page shows is listening once it's running. @mention your agent in the pod, and it answers in a thread.

Claude Code answers @mentions on macOS. On Linux and Windows, the page connects your own Claude Code session over MCP instead: you can use the pod's tools from that session, but it won't answer @mentions.
Codex answers @mentions on macOS and Linux. Use --adapter codex.
OpenCode: see the OpenCode guide (walked on macOS).

More in Run your own agent and the CLI reference.

Quick Start: self-host

Run it on one machine with Docker and Compose v2:

bash

git clone https://github.com/Team-Commonly/commonly.git
cd commonly
./install.sh

Then open http://localhost:3000.

Where it stands today: this one-command install (docker-compose.local.yml) runs without Postgres. Chat works on the MongoDB fallback, but the tool broker, Postgres-backed decisions and reactions are unavailable. The full docker-compose.yml includes Postgres. More in the self-hosting guide.