swing-mortem

What it does

The skill conducts prospective failure analysis (pre-mortem) to surface concrete, future failure scenarios across five categories (Technical, Organizational, External, Temporal, Assumption). It enforces exact, specific narratives with measurable triggers and outcomes, grounded in Gary Klein's pre-mortem technique. It requires enumerating five distinct scenarios (one per category), each with a concrete narrative, plausible justification, and a measurable consequence. It mandates a formal failure frame, a risk matrix selecting the top 3 risks by priority, leading indicators for those risks, circuit breakers with concrete triggers and fallbacks, and a single-sentence pre-mortem summary. The process is executed in six phases (Set the Failure Frame, Failure Scenario Generation, Likelihood x Impact Matrix, Leading Indicators, Circuit Breakers, Pre-Mortem Summary) and outputs a final formatted document with these sections.

How it works

• Establish the Failure Frame (Subject, Timeframe, and a complete failure statement).
• Generate exactly 5 failure scenarios, one per category (Technical, Organizational, External, Temporal, Assumption). Each scenario includes: What happened (2-4 sentences), Why it was plausible (1-2 sentences), and Concrete consequence (specific, measurable impact).
• Create a Risk Matrix rating Likelihood and Impact for each scenario, then compute Priority and select the top 3 by priority (ties broken by higher Likelihood).
• For each of the top 3 risks, define 2-3 Leading Indicators with explicit measures, thresholds, locations to observe, and lead time.
• For each top risk, define a Circuit Breaker with a concrete Trigger (threshold), a Fallback (specific alternative), Cost of delay, and Decision owner.
• Produce a Pre-Mortem Summary BLUF that synthesizes the highest risk, its leading indicator, and the escape hatch, plus cost implications and monitoring timeline.

When to use it

Use when you want a structured, adversarial forward-looking risk review for plans, architecture decisions, technology adoption, business strategy, or feature launches. Triggers to initiate the analysis include phrases such as "리스크", "위험", "실패하면", "swing-mortem", "뭐가 잘못될 수 있어", "risk", "what could go wrong", "걱정되는 점", "failure modes", "리스크 분석", or "위험 분석".

What it can touch

• Tools: Read, Grep, Glob, Bash, Agent
• Focus: It requires constructing narratives and measurable thresholds; it does not perform code edits or runtime actions unless those tools are invoked to fetch or inspect artifacts.

Caveats

• Requires precise, concrete scenarios with quantified details (no generic risks).
• Each category must contain exactly one scenario; no skipping categories.
• The swing-mortem summary is mandatory and must synthesize risk, leading indicator, and escape hatch in one paragraph.
• Assumes complete failure in framing; the analysis aims to surface explicit, testable thresholds and fallbacks.