What problem does it solve? After an epic finishes, no single coding session saw the whole change, so cross-story defects, god-class growth, and spec drift go unnoticed. This Skill reads the evidence an epic left behind — spec, story files, full diff range, per-story commits, sprint status — and produces a retrospective where every finding carries a source reference, ending in an acceptance verdict. ## Core Features & Use Cases - Evidence-gathering pipeline: Derives the diff range and per-story commit attribution via bundled git scripts, with explicit recording of missing evidence so readers can tell "checked and clean" from "never checked." - Aggregate analysis views: Detects architecture deltas, duplication, god-class growth, pattern divergence, and spec-to-implementation drift that per-commit review cannot see. - Acceptance verdict and sprint-status update: Renders accepted / accepted-with-open-items / rejected, writes a structured retrospective document, and safely updates sprint-status.yaml with action items through a comment-preserving, rollback-validated script. - Use Case: After an unattended multi-story epic run, ask for a retrospective to get a sourced findings report, routed action items with owners, and a machine-readable verdict an orchestrator can gate the next epic on. ## Quick Start Ask the AI to run a retrospective on the most recently completed epic and produce a sourced findings report with an acceptance verdict.
