Swing Mortem

Erzwingt ein Gary Klein pre-mortem: assume total failure, then find quantified risks and escape hatches.

von TheStack-ai · TheStack-ai/swing-skills

Getestet · Funktioniert ★ 9.6/10

Swing Mortem — Erzwingt ein Gary Klein pre-mortem: assume total failure, then find quantified risks and escape hatches.

Was es kann

Führt ein structured 6-phase pre-mortem on a plan, architecture, or launch: generates exactly 5 specific failure scenarios (technical, organizational, external, temporal, assumption), ranks them on a likelihood x impact matrix, and defines measurable leading indicators and circuit-breaker triggers for the top 3. Triggers on risk-analysis phrasing in English and Korean ('what could go wrong', '리스크 분석'); explicitly hands off to sibling skills for current-state review or option generation.

Testbericht

Führte ein real pre-mortem on this very testing batch aus and it correctly flagged the GitHub API rate-limit error we'd hit minutes earlier in-session as a quantified Technical failure scenario — the un-skilled baseline could only produce 'might have scaling issues' style generic bullets with no numbers or thresholds at all.

Getestet am: 2026-07-14 · Claude Code 2.x (agent harness)

Installation

git clone https://github.com/TheStack-ai/swing-skills
cd swing-skills
mkdir -p ~/.claude/skills
cp -r skills/swing-mortem ~/.claude/skills/swing-mortem

Befehle & Beispiel-Prompts

  • /swing-mortemErzwingt ein Gary Klein pre-mortem: assume total failure, then find quantified risks and escape hatches.

Skills reagieren auf normale Anfragen — keine Slash-Befehle nötig. Nach der Installation aktivieren Prompts wie diese den Skill (auf Englisch):

  • Before we commit to this migration plan, what could go wrong?
  • Do a pre-mortem on our Q3 launch plan with quantified risks.
  • Assume this rollout totally failed, what are the top risk scenarios?