Terminal
Terminal/Skill
Reviews your spec skeptically before any code is written, backs each finding with evidence, and ends with a go or no-go decision.
By Julius Brussee
Works with
npx skills add juliusbrussee/cavekit --skill reviewEvery finding cites evidence — file:line or a source. No evidence → flag [unverified]. Default to refuted: a flaw you cannot prove is a flaw you note, not one you wave through.
An LLM cannot self-correct on its own judgment — left alone it drifts or degrades. Review fixes that the only way that works: a separate skeptic anchored to an external oracle — the code, §R, the test suite, the docs. "Looks good" is not a review. A refutation attempt is.
/build on a high-blast-radius change (shared module, auth, data, money, public API).Skip for a trivial, reversible, well-understood change. Adversarial review on a typo hallucinates flaws & wastes the budget — the self-critique paradox is real.
Read the spec: §G §C §I §R §V §T. Hold the whole thing. You review the spec, not your memory of the conversation.
Build a reviewer with real authority, not a generic critic:
A reviewer with no evidence is just an opinion. Earn the authority first.
Attack the spec on these axes. For each, try to find the case where it breaks:
Each finding: evidence → claim → severity.
No evidence? Down-rank to NOTE & tag [unverified]. ⊥ inflate a hunch to BLOCK.
## review verdict
BLOCK: 1 — §I.api shape ≠ caller src/client.ts:40. fix §I before build.
HARDEN: 2 — drafted V8 (idempotent refund), V9 (tx around dual write).
NOTE: 1 — §T4 vague, split before /build.
gate: NO-GO until BLOCK cleared. then /build §T after spec writes V8,V9.
GO or NO-GO, never a shrug. Review is the checkpoint that stops a confident wrong build.
[unverified].