Cold-eye What it does
Cold-eye is an installable skill for AI agents. It checks whether a newcomer can understand and use what you are about to ship.
Give it a finished artifact. An agent reads the skill and writes a critique of missing instructions, contradictions, and unsupported claims.
Get started — install the skill in your agent, then use it.
You were in the room. The file looks finished. Locally coherent files can still fail as one product.
What it reads, writes, and changes
| Reads | A finished skill, spec, page, package, site, or repo |
| Writes | <name>.cold-eye.md next to a file, or cold-eye.md at a system root |
| Changes | Nothing, unless you separately ask for an edit |
| Verdict | Means |
|---|---|
| holds | A newcomer can run the job |
| close | A few edits to the file would close the gap |
| fails a hostile read | Too much still has to be invented |
The gate is the claim: done, or ready. A direction you might still change is out of scope.
This is a readiness pass. It is not a code audit, a security audit, or a comprehensive test run, unless those actions are the subject's own claimed checks.
The skill file uses hour for the procedure the subject tells someone to run: start, read, write, stop. Public pages say procedure or job for the same thing.
The critique shape.