What is atl?
atl is a command-line tool that installs teams of AI agents into a project — the same way npm installs packages of JavaScript code, or brew installs Unix binaries.
The problem
Using Claude Code well requires configuration: agents, skills, and rules that shape how the model reasons about your codebase. You end up copying files between projects, forking someone else's setup, and slowly watching them drift apart. Every new project re-solves problems that a previous project already solved.
The answer
A team is a package of agents, skills, and rules, built around a particular kind of work. One team might be geared for a .NET + Flutter + React stack with a Docker-compose production layout. Another might be for a Next.js + Sanity + Vercel blog stack. A third for data pipelines with Airflow and dbt.
atl install some-team fetches the team, caches it, and symlinks its agents, skills, and rules into your current project's .claude/ directory. Claude Code sees the team the moment you open the editor.
When the team author ships a fix, you run atl update and every project that uses that team picks up the change. Your projects stop drifting.
Not a fork
Teams can extend other teams. If you want 95 % of software-project-team but one different agent and one extra skill, you write a tiny team that extends it, overrides the one agent, and adds the skill. The parent keeps evolving in its own repo; your child inherits updates for free.
Not a walled garden
Every team is just a Git repository with a team.json file at the root. There's a public registry so teams get short names like software-project-team, but you can install from any Git URL. The schema is public. The CLI is MIT-licensed Go. The spec is documented here.
Who is this for?
- Developers who want a solid Claude Code setup without hand-rolling it for each project.
- Team leads who want to standardize how their company uses Claude across repos and onboard new engineers in minutes.
- Stack authors who want to publish opinionated agent teams the way framework authors publish CLIs today.
Where it stands
atl is at v0.1.x — early, but real. One team (software-project-team) ships a production-grade 13-agent bundle and is used to build at least one real product. The ecosystem is MIT-licensed and open for PRs.
Next up: