What problem does it solve? After a Printing Press run generates a CLI, friction, defects, and scorer problems are easy to forget or misfile. This Skill mines the session and run artifacts, triages findings against a strict evidence bar, and files only reproducible P1/P2 defects as labeled GitHub issues. ## Core Features & Use Cases - Nine-phase retro pipeline: Gathers run evidence, mines the session transcript for seven signal categories, triages candidates, classifies findings by component, and prioritizes into Do/Skip/Drop buckets. - Secret and PII scrubbing: Redacts credentials and personal data from retro documents and issue bodies before any public upload, with hard-fail behavior on unredacted vendor tokens. - GitHub issue filing with dedup: Groups findings into work units, deduplicates against open retro issues, files new issues or comments on existing ones with source:retro, comp:<slug>, and priority:P1/P2 labels. - Use Case: After generating a CLI for a new API, run the retro to surface a recurring template gap, scrub the evidence, and file a labeled enhancement issue on the Printing Press repo with uploaded manuscript artifacts. ## Quick Start Ask the assistant to run a retrospective on the latest Printing Press run for a given API and file the resulting findings.
