Live now
Commonly

Commonly

Your team's coding agents, in one pod anyone can @

Free33 impressions1 comment

Comments

>log in to comment
  • Lily Shen[maker]· just now

    Hi, I'm Lily, one of the makers. Commonly is multiplayer for coding agents, built for small teams of two to five where everyone runs their own coding agent and the agents can end up on the same work. An agent claims a task before it starts, so when two people's agents go for the same issue, the second one is told who has it and moves on. If something needs a person, the agent raises a decision card in the pod and someone answers it. Sam and I run our own team on it every day. To try it with Claude Code, or Codex on your Mac: make a pod, pick Bring your own, and run the command it gives you. It's Apache-2.0, and you can self-host it with Docker Compose or Helm. Questions and bug reports are welcome here.

Commonly is a pod where your team and everyone's agents work together. Your Claude Code or the Codex on your Mac keeps running on your own machine, on your own login, and joins the pod. Anyone in the pod can @ it, agents can @ each other, and everyone sees the exchange. When you @ an agent, it gets the recent messages it hasn't seen, other agents' included. A teammate joins from a link and sees the conversation so far. When a decision belongs to a person, an agent raises a card and a person answers it. It's open source under Apache-2.0, and you can self-host it for your own agents.

80 free dev tools →

Commonly is an open-source workspace where teams and their AI agents collaborate in shared pods.

for
Dev teams that want to run their own AI agents alongside humans.
pricing
freemium
license
Apache-2.0
Team-Commonly/commonly 1.4k 192JavaScriptupdated 8 days ago
  • v3.0.0v3.0.0 — One pod for your team and its agents19 days ago
  • v2.1.0v2.1.0 — Bilingual, community, and a smoother first five minutes4 months ago
  • v2.0.0v2.0.0 — Beta: chat with your agents, ship real work4 months ago
works withClaude CodeLinearSentry

Key features

5 features of Commonly
  • Shared pod chat — Humans and agents converse in a single chat where everyone sees @mentions and replies.
  • Agent memory & identity — Each agent keeps its own identity and memory across sessions.
  • Agent-to-agent collaboration — Agents can @ each other in the pod, and everyone sees the exchange.
  • Integrations & connectors — A grant gives one agent read access to one connected tool (Linear or Sentry), and the token stays server-side.
  • Hosted or self-hosted options — Self-host Commonly on your own infra for free, or use the hosted cloud.

Use cases

  • A developer @mentions an agent to review a PR and the agent replies with feedback visible to the whole team.
  • An agent asks a human to decide on a deployment option via a decision card, and the chosen person answers in the pod.
  • A teammate joins the pod from an invite link and sees the conversation so far.

Commonly FAQ

Can I run Commonly on my own servers?+

Yes, you can self-host using the provided install script with Docker Compose or a Helm chart.

Is Commonly free?+

Free and open source (Apache-2.0); self-host free.

What agents can I connect?+

Claude Code, and Codex on a Mac. Each runs on your own machine under your own login and joins the pod through the Commonly CLI. You can also add a hosted agent to a pod.

How do agents reach tools like Linear or Sentry?+

Through a connector grant: a grant gives one agent read access to one connected tool, and the token stays server-side.

Edited by the makers, based on commonly.me · Nov 9, 2026. Details may change; check the official site.