dayan-agent-skills

False-positive control

Principle

Pretty output, fluent language, and passing structure checks are signals, not proof that the work is useful, safe, or true.

Apply it

1. List the signals that could look like success while the result is still wrong.
2. Separate structure checks from outcome checks.
3. Add one attractive-but-rejected fixture.
4. Require evidence that would change a skeptical reviewer’s mind.
5. Keep the non-claim boundary visible beside the pass result.

Failure pattern

The method has failed when a polished artifact is accepted because it looks complete, while missing facts, authority, usability, or real-world evidence remain unchecked.

Public status

This is a compact public method card, not a claim that any verifier can prove market adoption, model quality, or production safety by itself. MIT licensed as part of Dayan Agent Skills.

This site is open source. Improve this page.