cat:run-retrospective

Automate retrospective analysis to identify mistake patterns and derive action items.

101|2|Updated Jan 7, 2026

Search Tags:#workflow #patterns #retrospective #action-items #learn-from-mistakes #escalations

Ready to run

npx skills add https://github.com/cowwoc/cat --skill cat-run-retrospective

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cat:run-retrospective
Source: https://github.com/cowwoc/cat/tree/main/plugin/skills/run-retrospective
Command: npx skills add https://github.com/cowwoc/cat --skill cat-run-retrospective

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates retrospective analysis to identify patterns in mistakes, derive actionable items, and track their effectiveness over time.

Core Features & Use Cases

Automated, scheduled retrospectives that run against accumulated mistakes.
Pattern discovery, action-item generation, and escalation of ineffective fixes.
Retrospective record creation and progress tracking integrated with learn-from-mistakes workflows.

Quick Start

Ensure the environment includes the required tooling (jq and shell access to the retrospectives store at .claude/cat/retrospectives).
Trigger a retrospective with the /cat:run-retrospective command or via thresholds defined in your configuration.
Review findings and implement action items; monitor effectiveness across subsequent retrospectives.

Frequently Asked Questions about cat:run-retrospective

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

FAQPage Schema

How do I automate scheduled retrospectives to identify patterns in mistakes?▼

Automated retrospectives identify mistake patterns by aggregating accumulated errors, classifying them by category, and generating action items. They run automatically via time- or count-based triggers against a retrospective data store.

How does a retrospective workflow derive action items from past mistakes?▼

Retrospective workflows derive action items by analyzing aggregated mistakes, discovering recurring patterns, classifying them by category, and producing actionable resolutions tracked for effectiveness over subsequent iterations.

What tools do I need to run automated retrospectives in my environment?▼

Running automated retrospectives requires shell access, the jq tool, and a configured retrospective data store located at .claude/cat/retrospectives to aggregate mistakes and generate records.

When should an escalation be triggered during a retrospective process?▼

Escalations trigger during retrospectives when previously implemented fixes fail to resolve identified mistake patterns. The workflow applies escalation templates to flag ineffective action items for further review.

Can I trigger a retrospective automatically based on time or count thresholds?▼

Retrospectives can be triggered automatically using configured time-based or count-based thresholds. They also support manual invocation via the /cat:run-retrospective command for immediate post-milestone reviews.

What is the best way to track the effectiveness of action items over time?▼

Tracking action item effectiveness involves monitoring implemented fixes across subsequent scheduled retrospectives. The system evaluates whether mistake patterns reoccur and generates escalations for ineffective resolutions.