crux-skill-memory-meditation-review

You can audit AI-generated research outputs by applying a structured adversarial review process that checks evidence, consistency, and completeness. This validates citation integrity and detects drift before finalization.

An adversarial review process for meditation agents is a quality gate that detects missing evidence, inconsistencies, and incomplete report requirements. It classifies findings by severity and applies controlled fixes or escalations.

To validate citation integrity and report readiness, apply a 13-dimension review framework that audits comprehensiveness, peer review quality, and finalisation requirements, checking accepted enhancements for finalisation.

Yes, content review supports bounded iteration controls and structured user-input escalation. It applies unambiguous fixes based on severity classification and enforces respawn protocol compliance for missing sections.

Handle incomplete report generation by validating report-generation readiness and coordinating deterministic respawn requirements for missing sections or visualizations to ensure reliable finalisation.

Adversarial quality gates are not suitable when lacking defined review dimensions and severity classification parameters, as the process requires structured iteration controls and respawn protocol compliance to function reliably.