retro

Generate engineering retrospective reports from git history and Claude session summaries.

Updated Apr 15, 2026
One-click install
npx skills add https://github.com/ToanPV90/dotfiles --skill retro-toanpv90
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retro
Source: https://github.com/ToanPV90/dotfiles/tree/main/claude/.claude/skills/retro
Command: npx skills add https://github.com/ToanPV90/dotfiles --skill retro-toanpv90

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you understand what you actually shipped and how your engineering work is trending, by converting git history plus recent session context into a structured retrospective.

Core Features & Use Cases

  • Quantitative work analysis: Produces commit counts, files changed, and lines added/removed from a selectable time window (including compare mode).
  • Quality signals & hotspots: Estimates test-file ratio and highlights the most-changed files and directories to reveal focus areas and risk areas.
  • Qualitative cross-reference: Pulls matching Claude session summaries to contextualize metrics with decisions, goals, and outcomes.
  • Use case: Run a weekly retro to answer questions like “what did I ship?”, “what were my main hotspots?”, and “did work patterns shift this week?”

Quick Start

Ask for a report by saying: “Run /retro for the last 7 days and include a comparison to the previous week.”

Frequently Asked Questions about retro

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an engineering retrospective report from git commit history?

To generate an engineering retrospective, you can run a weekly or ad-hoc review that analyzes git commit history and correlates it with recent Claude session summaries to produce an immutable markdown report.

Can I compare git commits and test-file ratio across different time windows?

Yes, you can use compare mode to measure changes in commits, lines, hotspots, and test-file ratio against a prior time window to see if your engineering work patterns have shifted.

How do I track engineering hotspots and identify heavily modified files?

Engineering hotspots are tracked by computing git metrics for commits, files, and lines changed, then highlighting the most-changed files and directories to reveal focus and risk areas.

Does generating a session summary retro require a specific repository setup?

Generating a retrospective requires git repository validation to compute commit metrics, ensuring the analysis is based on valid version-controlled data before saving the report under ~/.claude/retros/.

What is the best way to quantify shipped work and quality signals for a weekly review?

The best way to quantify weekly shipped work is by analyzing git history to produce commit counts, files changed, lines added or removed, and estimating the test-file ratio for quality signals.