What I'm building

Bookbag.ai
Run AI development. Keep control.

Coding agents are getting good at writing software. Bookbag is the layer around them: what they know, what they may touch, how work moves between them, who approves it, and the record of what happened. Built for engineering leaders and platform teams.

Brief
Assign
Work
Check
Approve
Record

Start free Talk to us Watch the experiments

The belief behind it.

Agents are becoming the workforce.

They pick up tasks, change code and open pull requests on their own, and teams are adding them faster than any hire.

A prompt is not a control system.

A rule written into an agent's instructions is advice. Until something outside the model enforces it, nothing is guaranteed.

So the controls live in the platform.

Routing, permissions, gates, approvals and the audit trail are enforced by Bookbag, not trusted to the agent.

The products

One account. Two halves of the problem.

Codebook organizes agents into factories that plan, build and review. Madebook governs their work where it merges.

Agent factories

Codebook

Lay out the work as assembly lines. Put an agent, a person, or an agent gated by a person on every station. Routing, retries, budgets and permissions are enforced by the platform, not trusted to the prompt.

codebook.work Coding governance

Madebook

Sits above every coding agent your team runs. Briefs each one from what is currently true, then posts one required check on every pull request. Agents that don't report don't merge.

madebook.ai

Who it's for

Teams moving from coding assistants
to agents that act on their own.

If your engineers already run several agents, these are the questions Bookbag answers.

Claude Code
Codex
Cursor
GitHub Copilot
Devin
Kiro
GitLab CI/CD
CircleCI

The questions Bookbag is built to answer

01Which agent is working on what?
02What is it allowed to touch?
03Who approved this change?
04Why did it make that decision?
05What happens when two agents conflict?
06Can we prove what happened six months from now?