$ agent-lab product
How it works
Docker sandboxes so coding agents can run without restrictions, safely. What Agent Lab does, what it’s built with, and how far along it is.
The premise
Agent Lab lets Codex and Claude Code run in their no-restrictions modes while Docker keeps them inside the one project folder you mount. If something goes wrong, the damage stays in the container.
Logins for the agent CLIs, GitHub and Google Cloud are kept in Docker named volumes. They carry over between sessions but never show up in the sandbox’s files. A single install script sets up a new machine.
What’s included
Unrestricted agent modes, limited to the project folder you mount
Credentials kept in Docker named volumes, outside the sandbox’s files
One install script, and images build the first time you use them
The same commands for logins and MCP servers across both agent CLIs
How it’s built
Agent Lab is built on Docker, Shell, PowerShell, Claude Code and Codex.
The repository is 44% Shell, 27% JavaScript, 24% PowerShell, 5% Dockerfile by volume.
It carries 37 commits since Aug 2026, most recently changed in Aug 2026.
How the work is done
Every project here starts as a written spec, and the spec stays the source of truth once there’s code. The repository keeps a build plan with current status, an append-only decision log of where the code departed from the plan, and a handoff document that describes where things stand, including the unfinished parts.
That’s how one person keeps this many projects moving, and it’s what this page is written from.
Current state
In development. Being worked on now.
The source is public, so you can check anything on this page against it.
Next
Put Agent Lab in your pipeline
Ask about hooking it up to the repositories and agents you already run.