retrospective

Analyze post-shipment data to extract learnings and propose CLAUDE.md updates.

3|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/cardinalconseils/claude-starter --skill retrospective-cardinalconseils
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retrospective
Source: https://github.com/cardinalconseils/claude-starter/tree/main/skills/retrospective
Command: npx skills add https://github.com/cardinalconseils/claude-starter --skill retrospective-cardinalconseils

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Retrospective automates post-ship analysis to extract learnings, patterns, and conventions, enabling continuous improvement across cycles.

Core Features & Use Cases

  • Automated extraction of learnings from completed ships to inform future planning and CLAUDE.md updates
  • Velocity metrics, hotspot detection, and convention tracking to guide process improvements
  • Proposes CLAUDE.md updates and stores insights under .learnings/ for auditability

Quick Start

After shipping, trigger the retrospective to generate learnings, update CLAUDE.md proposals, and save results to .learnings/.

Frequently Asked Questions about retrospective

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

FAQPage Schema
How do I automate post-shipment retrospective analysis to extract learnings?

To automate post-shipment retrospective analysis, trigger the skill after shipping to read PRD artifacts, session logs, and observability data. It automatically extracts learnings and patterns to generate insights and CLAUDE.md updates.

What is the best way to generate CLAUDE.md updates from completed ships?

The best way to generate CLAUDE.md updates is running an automated retrospective that reads completed ship artifacts. It proposes conventions and stores insights under the .learnings/ directory for auditability and future planning.

How do I track velocity metrics and hotspots after shipping code?

You can track velocity metrics and hotspots after shipping by applying automated retrospective analysis. It evaluates post-shipment data and observability logs to detect hotspots, track conventions, and guide process improvements.

Does the retrospective process require pre-existing PRD artifacts and session logs?

Yes, the retrospective process requires pre-existing PRD artifacts and session logs. It reads these artifacts alongside observability data after shipping to extract accurate learnings and generate meaningful convention proposals.

Can I run retrospective analysis in interactive mode to review learnings before saving?

Yes, you can run retrospective analysis in interactive mode. Alongside the automated mode, interactive mode allows you to review extracted learnings and observability insights before saving results to .learnings/session-log.md.