The NSFW rule now tells the model to inspect the actual image for adult
content and flag it even when the bio text is innocuous, explicitly
overriding the text-first guidance.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Update description; flag NSFW and crypto as signals
Joseph Antolick+1: Claude Opus 4.8
ff18b7e
New longer description blurb. Prompt now always adds an "NSFW" signal (and
notes it in the rationale) for adult content, and a "crypto" signal for
crypto/web3 mentions, placed first so the signal cap can't drop them.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Rename Crankycheck -> TemperCheck
Joseph Antolick+1: Claude Opus 4.8
aa00509
Package crankycheck/ -> tempercheck/, CrankyVerdict -> TemperVerdict,
CRANKY_* env vars -> TEMPER_*, plus all UI copy, prompt branding, README
Space card, and docs. Behavior unchanged; tests pass.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>