What problem does it solve? Engineering teams lack visibility into what actually shipped each week, how work patterns affect output, and whether code quality is trending up or down. This Skill turns raw git history into a structured retrospective with metrics, per-person breakdowns, and historical trend tracking. ## Core Features & Use Cases - Commit and Metrics Analysis: Aggregates commits, PRs, insertions/deletions, test LOC ratio, and active days over a configurable window (24h, 7d, 14d, 30d). - Team-Aware Breakdowns: Identifies the current user via git config and produces a per-author leaderboard with praise and growth areas for each contributor. - Work Pattern Detection: Builds hourly commit histograms, detects coding sessions using a 45-minute gap threshold, and flags late-night clusters and fix-ratio spikes. - Comparison and Global Modes: Compares the current window against the prior period, or runs a cross-project retro across all AI coding tools. - Use Case: At the end of a sprint, run the retro to see that fixes made up 54% of commits, two files churned 8+ times, and your deep-focus sessions dropped versus last week, then act on those signals. ## Quick Start Ask the assistant to run a weekly engineering retrospective over the last 7 days of this repository's commit history.
