Everything you need to run agents at work.
Chat, sandboxes, a console, and full tracing — plus a CLI and a Terraform provider. All of it on infrastructure you control.
Chat
Where your team works with agents. Ask in a thread, attach the files, get an answer back with the chart already in it.
agents wired to your systems·works while you’re away
Explore Chat →Sandboxes
Run Claude Code, Codex, or your own agent in a real terminal — with git and the rest of your engineering tools around it.
claude code·codex·syncs with your editor
Explore Sandboxes →Tracing
A full record of what an agent did and why — every tool it called, every prompt it saw, every attempt that was blocked.
every tool call·the exact prompt·watch it live
Explore Tracing →Console
Decide what your agents are allowed to run, what they can reach, what they may spend, and who gets to use them.
who can use what·where agents may connect·spend per team
Explore Console →CLI
Everything the apps do, scriptable. The same command works from your laptop and from inside a running agent.
scriptable·runs in CI·your agents can call it
Explore CLI →Terraform
Define agents and what they’re allowed to touch in code. Granting access becomes a pull request someone approves.
agents as code·review before apply
Explore Terraform →See it on your own cluster.
Thirty minutes, your infrastructure, your stack. Or skip the call — it is one Helm release onto a cluster you already run.