What problem does it solve? Research artifacts often pass structural checks while still containing weak evidence, unfalsifiable claims, or incoherent arguments. This Skill performs a semantic peer review of an Agent-Native Research Artifact (ARA) so authors get an objective, calibrated epistemic critique before publication or release. ## Core Features & Use Cases - Six-Dimension Scoring: Scores evidence relevance, falsifiability, scope calibration, argument coherence, exploration integrity, and methodological rigor on a 1-5 scale with documented anchors. - Severity-Ranked Findings: Produces critical/major/minor/suggestion findings with verbatim evidence spans, reasoning, and constructive fixes, then maps the mean score to a Strong Accept-to-Reject grade. - Structured Report Output: Writes a complete level2_report.json at the artifact root, including per-dimension strengths, weaknesses, questions for authors, and the read order. - Use Case: After Level 1 structural validation passes on an ARA containing claims, experiments, and an exploration tree, run this review to catch a causal claim supported only by correlational evidence before the artifact is published. ## Quick Start Review the ARA in my artifact directory with the rigor reviewer and generate the level2_report.json epistemic assessment.
