with one click
apr-model-qa-playbook
apr-model-qa-playbook contains 3 collected skills from paiml, with repository-level occupation coverage and site-owned skill detail pages.
Skills in this repository
Systematic coverage gap analysis and test writing for the APR Model QA Playbook. Uses pmat query --coverage-gaps for highest-ROI targets, provides test patterns for each crate (proptest for gen, Evidence assertions for runner, MQS scoring for report), handles CommandRunner trait extensions, clippy pedantic/nursery landmines, and ExecutionConfig construction. Target: 95% library coverage.
Diagnoses and fixes G0-G4 gateway failures in model qualification runs. Interprets MQS zero-scores, evidence JSON, and stderr output. Covers LAYOUT-002 violations, config.json mismatches, garbage detection, crash analysis, and all 5 contract invariants (I-1 through I-5). Includes the GarbageOracle detection patterns and conversion diagnostics.
Guides end-to-end HuggingFace model certification: playbook creation from templates, running qualification at any tier (Smoke/MVP/Quick/Standard/Deep), collecting evidence, computing MQS scores with G0-G4 gateways, updating models.csv, and syncing the README certification table. Covers SafeTensors ground truth, LAYOUT-002 compliance, and Popperian falsification methodology.