What problem does it solve? Infrahub repositories can silently accumulate broken schema relationships, unregistered Python components, deprecated fields, and mismatched query names that only surface as runtime failures during sync or pipeline execution. This Skill runs a nine-phase audit that catches these issues before deployment and reports them with severity levels and concrete fixes. ## Core Features & Use Cases - Nine-phase audit procedure: Validates project structure, schemas, object data, Python checks/generators/transforms, cross-references, registration completeness, best practices, deployment readiness, and YAGNI cost-to-fix findings. - Structured severity-ranked report: Writes findings to AUDIT_REPORT.md organized by CRITICAL, HIGH, MEDIUM, LOW, and INFO severity, each with file location and suggested fix. - Deprecated field migration: Detects deprecated display_labels usage and generates the exact display_label Jinja2 replacement string. - Use Case: Before deploying an Infrahub repository or after a major refactor, run the audit to catch a generator missing allow_upsert=True, a relationship identifier mismatch, or an orphan check file that would otherwise fail silently in production. ## Quick Start Ask the assistant to audit this Infrahub repo and it will walk the nine phases and write the findings to AUDIT_REPORT.md.
