fbhadha/skills (adversarial-review-engine)

Two modes of ruthless critical analysis. Prosecution mode dismantles someone else's writing, argument, or claim as a hostile peer-reviewer and opposing counsel — auditing every fact, quote, citation, and statistic against primary sources, delivering a severity-ranked Kill List, steelman counter-narrative, verdicts, and concessions. Devil's-advocate mode turns the same weapons on the user's own position, draft, or point of view to make it more robust — surfacing the strongest objections an opponent would raise, then hardening the position and restating it at full strength. Use when the user wants a text torn apart ("tear this apart," "fact-check this," "find the holes," "is this legit") or their own thinking challenged ("play devil's advocate," "challenge my position," "stress-test my argument," "poke holes in my thinking," "make this more robust," "what would critics say"). Trigger even when the user doesn't name the skill.

Claude Skills are markdown + scripts loaded into Claude Code; scanned with the same pipeline.

Source:

github.com/fbhadha/skills

Provenance:

Found by crawling (provenance, not a security rating)

We found this by crawling public sources and nobody has attested to it. That is how most of the registry gets here, and it is not a mark against the artifact — it only means no authenticated act of publication is on record. An artifact can be the vendor's own and still appear here.

1 critical finding reported by cisco-skill-scanner. The composite score alone does not reflect this — read the per-scanner reports below.