retrospective-counterfactual

Reconstruct timelines and classify causes by knowability for incident retrospectives.

7|2|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/AndurilCode/craftwork --skill retrospective-counterfactual
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retrospective-counterfactual
Source: https://github.com/AndurilCode/craftwork/tree/main/skills/retrospective-counterfactual
Command: npx skills add https://github.com/AndurilCode/craftwork --skill retrospective-counterfactual

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you rigorously understand why an event went wrong—or why it unexpectedly succeeded—so you can learn actionable improvements instead of relying on blame or hindsight.

Core Features & Use Cases

  • Post-mortem retrospectives: Reconstruct timelines, identify immediate causes, enabling conditions, root causes, and contributing factors to find systemic fixes.
  • Counterfactual analysis: Evaluate “what would have happened if we’d done X differently” to assess decision quality and avoid survivorship bias.
  • Knowability classification: Classify causes as predictable/detectable/preventable to prioritize the right type of corrective action (process, monitoring, resilience).

Use cases: incident reviews, failed launches, debriefs after project delivery, and after-the-fact evaluation of team or system decisions.

Quick Start

Use the retrospective-counterfactual skill to analyze the incident by reconstructing a timeline, classifying causes by knowability, and proposing 2–3 owned corrective actions with verification criteria.

Frequently Asked Questions about retrospective-counterfactual

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

FAQPage Schema
How do I run a root cause analysis for an incident review without relying on hindsight bias?

Root cause analysis avoids hindsight bias by reconstructing a structured timeline and classifying causes by knowability—predictable, detectable, or preventable—to identify systemic fixes rather than assigning blame.

What is counterfactual reasoning in post-mortem retrospectives?

Counterfactual reasoning in post-mortems evaluates what would have happened if a single decision varied, preserving causal plausibility to assess decision quality and prevent survivorship bias in incident reviews.

How do I conduct a debrief after a failed project launch?

Conduct a debrief after a failed launch by reconstructing the event timeline, identifying immediate causes and enabling conditions, and proposing owned corrective actions with clear verification criteria.

When should I apply survivorship-bias-aware reviews to successful outcomes?

Survivorship-bias-aware reviews apply to successful outcomes when you need to understand why the result occurred unexpectedly, ensuring you do not incorrectly validate flawed decisions that happened to succeed.

What is the best way to classify causes during an incident review?

The best way to classify causes during an incident review is to categorize them by knowability—predictable, detectable, or preventable—to prioritize the correct type of corrective action, such as process changes or monitoring.