What problem does it solve? AI-generated content often contains fabricated citations, invented statistics, and plausible-sounding errors that are hard to catch manually. This Skill systematically extracts every verifiable claim from AI output, finds independent sources, and flags hallucination patterns before you rely on the content. ## Core Features & Use Cases - Three-Layer Verification Pipeline: Extracts and categorizes claims (factual, statistical, citation, entity, causal, temporal), searches the web for supporting or contradicting sources, then runs an adversarial review targeting common hallucination patterns. - Two Operating Modes: Active mode embeds inline verification summaries into every substantive response, while one-shot mode runs a full verification on a specific text you provide. - Structured Verification Reports: Produces reports using a template with confidence ratings (VERIFIED, PLAUSIBLE, UNVERIFIED, DISPUTED, FABRICATION RISK), source links, and flagged items requiring attention. - Use Case: After receiving a legal analysis citing several court cases, invoke the skill to confirm each citation actually exists, check that holdings are described accurately, and receive a report highlighting any fabricated references. ## Quick Start Ask the AI to doublecheck the previous response and produce a full verification report with sources.
