← ClaudeAtlas

premortemlisted

Run a premortem on a plan, launch, product, hire, strategy, partnership, pricing change, or high-stakes decision by assuming it failed 6 months from now and working backward to identify why. Use when the user says "premortem this", "premortem my", "run a premortem", "what could kill this", "future-proof this", "stress test this plan", "what am I missing here", "find the blind spots", "what could go wrong", "am I missing anything", "poke holes in this", "where will this break", or "devil's advocate this" in relation to a concrete plan or commitment. Do not use for simple factual questions, ordinary draft feedback, vague ideas with no plan yet, or LLM Council requests.
0xAddict/threadwork · ★ 0 · AI & Automation · score 70
Install: claude install-skill 0xAddict/threadwork
# Premortem Run a prospective-hindsight failure analysis: assume the plan has already failed 6 months from now, then work backward to identify why, what assumptions enabled the failure, and how to revise the plan before execution. ## Use When Use this skill for concrete, reversible plans or commitments where the cost of being wrong is meaningful: - Product or feature launches - Pricing or business model changes - Hiring decisions - Strategic pivots or positioning changes - Partnerships, deals, or major operational commitments - Any launch, plan, or decision the user asks to stress test for blind spots Do not use this skill for: - Vague ideas with no concrete plan yet; help define the plan first - Questions with one correct factual answer - Creative editing or ordinary feedback on a draft - Decisions that are already irreversible - LLM Council requests ## Context Minimum Before running the premortem, make sure you know: 1. **What it is**: the plan, product, launch, hire, strategy, or decision in one sentence. 2. **Who it affects**: customer, audience, team, buyer, stakeholder, or user. 3. **What success means**: the outcome the user wants. First scan existing context: - Current conversation - `CLAUDE.md` or `claude.md` - `memory/` - Files explicitly referenced by the user - Nearby briefs, plans, docs, or project files that appear relevant Keep this scan quick, roughly 30 seconds. Prefer `rg` and `rg --files`. If any of the three minimum context items are missing,