ai-output-validator

Hallucination detection analyzes AI outputs for factual inaccuracies and unsupported claims. This Skill cross-references content against external sources and flags statements that lack factual basis, returning a validation report with identified errors and confidence scores.

Yes. This Skill performs automated fact-checking by validating AI outputs against external sources, scoring accuracy, and identifying claims that need verification or correction with specific improvement recommendations.

Safety validation screens for bias, harmful content, and logical inconsistencies in AI outputs. The Skill applies multi-criteria scoring across hallucination, bias, and consistency dimensions, returning actionable fixes prioritized by severity.

Consistency checking evaluates whether AI outputs maintain logical coherence and factual alignment throughout. This Skill applies logicality checks and completeness scoring to identify contradictions and gaps, with concrete suggestions for improvement.

Yes. The Skill validates AI outputs across LLM responses, documentation, technical articles, and chat interactions. It applies the same fact-checking, consistency, and safety checks regardless of source format.

Validation returns a comprehensive report including quality scores, identified errors categorized by type, bias and safety flags, completeness assessment, and prioritized actionable recommendations for correction.