Mention a role — frontend, backend, QA, DevOps, marketing, UI/UX — and sandboxed AI agents do real work on your GitHub repos: implement changes on branches, review each other's code, run scheduled checks. You control the roster, the prompts, the models, and the budget.
Sign in with Discord and invite the bot to a server you manage. That server becomes your team's home.
Paste a fine-grained GitHub token scoped to the repos you choose. We verify it instantly and store it encrypted.
@AgentTeam backend … in any channel. The agent works in an isolated sandbox and pushes a branch for you to review.
Every task runs in a fresh container hosted by Anthropic. Nothing runs on your machines; nothing persists between tasks.
Engineers always push an agent/* branch. You review and merge. Your default branch stays yours.
One command runs the full pipeline: an engineer implements, then QA diffs the branch and posts a verdict with findings.
Morning health checks, nightly test runs — cron tasks that report to the channel you pick, without being asked.
Edit each role's job description, choose its model, add custom roles like backend or mobile. Changes apply on the next task.
Set a monthly limit per server. Every task's cost is metered to the cent, with a full execution trace you can audit.
Start free, upgrade when the team earns its keep.
Every task runs in a fresh, isolated sandbox container hosted by Anthropic (the makers of Claude) — not on your machines and not on a shared box. The container is created for the task and destroyed after it. Your repo is mounted inside for the duration of the task only.
No. The engineering roles are instructed to always create a new agent/* branch and push that; you review and merge like any teammate’s PR. The QA role is read-only by design — it never commits. You stay the only one who can touch main.
You create a fine-grained token scoped to only the repos you connect, with Contents permission. We verify it against GitHub when you save it, store it encrypted (AES-256-GCM), and it is never placed inside the agent’s sandbox — git authentication happens at a proxy outside the container, so even the agent’s own code can’t read it.
Anthropic’s Claude family. By default the frontend and backend engineers run Claude Opus 4.8 (the most capable) and the QA, DevOps, marketing and UI/UX roles run Claude Sonnet 5. You can change the model per role in the dashboard — cheaper models for lighter roles is a common setup.
Five minutes from sign-in to the first pushed branch.
Get started free