I built a lie detector for AI coding agents. Then I read the rules.

When an AI agent commits a change to code, it makes a claim about what it changed: "fixed typos" or "removed excessive whitespace".

In situations where one agent makes a change and one human reviews it, this is fine.

But in situations where many agents make changes at once, a human simply can't verify all the changes in the codebase.

Every agent that's pushing a change to the codebase must submit a claim about what it changed. The platform will verify that the claim matches what actually changed before merging the code.

Five agents push at once. Verdicts arrive in seconds, each with evidence. Recorded on a live deployment; the demo is now shut down to avoid costs.

The system passed 5 full demo runs, 28 agents each, in which they all were pushing together.

In every demo, the system correctly rejected 11 agents that misreported their claims, and 0 agents that were honest.

Every verdict was provided with supporting evidence, and typically 6-7.5 seconds of median time to solution.

I only read the competition rules properly a week into building it, and I'll have more to say on this later.

Agents can describe their changes any way they like. claimcheck checks every description against the real diff, and only verified changes merge.

In 5 full demo runs in a row (28 agents per run, pushing at the same time), claimcheck rejected all 11 agents that misreported, in every run, and wrongly rejected 0 honest claims. Each verdict came with evidence, in a median of 6–7.5 seconds. (Honest changes that touch a file another verified change already edited are blocked on purpose: the first verified change wins.)

A Git platform layer for AI agents, built on Cloudflare Workers and Artifacts. Each agent works in its own fork and must push a claim: which files it changed and what each change does. Entry for the Cloudflare "build the next Git platform" competition. License: MIT.

An AI agent's commit message is a claim, and nobody checks it. An…