What problem does it solve? Actuarial models built with gaspatchio can contain hidden anti-patterns (like map_elements or Python loops) that destroy performance, and methodology errors (wrong decrement timing, misaligned lookup tables) that silently produce wrong numbers. This Skill provides a structured, evidence-based review process that catches both code quality and actuarial methodology issues before models reach production. ## Core Features & Use Cases - Two-layer review: Layer 1 checks gaspatchio code quality against a catalog of 10 anti-patterns; Layer 2 applies an ASOP 56-informed checklist covering correctness, assumption integrity, change impact, and documentation. - Severity-gated completion: Issues are classified as Critical, Important, or Minor, and the review is blocked until all Critical and Important findings are resolved with file:line references. - Executable verification: Runs the model via gspio run-single-policy and inspects parquet output with gspio describe --json rather than trusting stdout or claims that results "look reasonable". - Use Case: After building a term life projection model, invoke this Skill to verify formulas, lookup table mappings, and decrement ordering, then receive a structured review report with actionable fixes before merging to production. ## Quick Start Review my gaspatchio model in model.py for code quality and actuarial methodology issues, then give me a severity-classified report.
