verify-before-claiming
Before reporting a task as done, run the check that would fail if it were not — an explanation is not a fix.
Conferida por quem revisou e leu o cartão, não reivindicada pelo próprio arquivo. Um cartão que o agente destila durante uma execução que consumiu conteúdo não confiável nasce contaminado e fica retido para revisão antes de alguma vez ser recuperado.
Quando ele vem à mente
- finished the change
- about to report success
- the fix looks right
- summarising what was done
Evite e Verifique são onde costuma estar o valor. Faça é a seção que todo mundo escreve.
Gatilho
You are about to say a task is complete. Anything from "fixed the bug" to "updated the config" to "the tests should pass now".
It does not apply when the task was genuinely to explain, review or investigate something. Those finish with an answer, and demanding a diff from them is its own mistake.
Faça
- Name the observable that would be different if the work had happened. A file whose contents changed, a command whose exit code flipped, a row that now exists.
- Check that observable. Actually run the command, actually read the file back.
- Report what you saw, including the command and its output — not your expectation of it.
- If nothing observable changed, say that plainly instead of describing the change you intended.
Evite
Reporting the plan as the outcome. The failure reads like: "I updated the handler to check the token before branching" — fluent, specific, technically accurate about the intent, and describing an edit that was never written to disk.
It is tempting because a convincing explanation feels like evidence. It is not. The explanation is produced by the same process that would produce it whether or not the edit landed, so it carries no information about whether it did.
Also avoid checking something adjacent to the claim: running the full suite proves the suite passes, which is not the same as proving this change did this thing. Pick the check that would have failed before.
Verifique
The claim and the evidence describe the same event, and the evidence came from the machine.
Concretely: a diff that is non-empty, a test that failed before and passes now, output pasted
rather than paraphrased. If you cannot produce one, the honest report is "I could not verify this",
which is a useful thing to say and takes one sentence.
Risco
Over-applied, this turns a two-line documentation fix into a ceremony, and there are tasks whose result genuinely is prose. The cost of the check should stay well under the cost of being wrong.
The subtler risk: a check that always passes is worse than no check, because it launders a guess into a verified claim. If the verification cannot fail, it is not verifying anything.
Como usar
O cartão é dado. Clone o repositório e importe pelo caminho — o que chega pela rede é tratado como contaminado e fica retido para aprovação, que é o comportamento desejável e a razão de não haver um instalador de uma linha aqui.
git clone https://github.com/brcampidelli/chimera-agent.gitchimera skills-import chimera-agent/skills/verify-before-claiming/SKILL.mdIntegridade
SHA-256 do arquivo como publicado. Quem importa pode conferir que o que recebeu é o que esta página mostrou.
c3c90d1467f39a63fb73dc21f2ac4aaf176306e532a3c58954dcbbc85566c2f5