How to Pre-Mortem Any Decision With AI: The Future Failure Audit Skill for Claude 🔮
The highest-leverage Claude skill most founders are missing. It assumes your plan has already failed, finds the hidden assumption that kills it, and rewrites it before you commit a single dollar.
👋 Hey, Linas here! Every day, I break down 3 stories shaping the future of FinTech & Artificial Intelligence - plus the money movements and trends worth tracking. First time here? 395k+ FinTech and AI leaders get this daily. Join them:
Most expensive mistakes don’t start as obviously bad ideas. They start as plausible plans that nobody stress-tested while there was still time to change them.
The launch that looked ready. The hire who interviewed perfectly. The round that should have closed. The investment memo that read well right up until it didn’t.
A pre-mortem fixes this.
Instead of asking “what could go wrong?”, it assumes the plan has already failed and works backward to explain exactly how it died, while you can still do something about it.
The technique rests on a documented cognitive effect called prospective hindsight: research by Deborah Mitchell, Jay Russo, and Nancy Pennington found that imagining a future event has already happened makes people roughly 30% better at identifying the reasons it would (Mitchell et al., 1989; Klein, HBR 2007). Nobel laureate in economics Daniel Kahneman later endorsed the pre-mortem as one of the few practical defenses against overconfidence, groupthink, and the planning fallacy.
The problem is that almost nobody runs one. Pre-mortems are awkward to organize, easy to skip, and quietly threatening to a team that has already fallen in love with its plan. The single highest-return decision tool in behavioral science thus sits unused while preventable failures stack up.
This guide therefore turns the pre-mortem into something you can run in 60 seconds, on any decision, indefinitely: a packaged Future Failure Audit skill for Claude. Upload it once, and Claude becomes a structured adversary on demand. It reconstructs your plan, generates specific failure modes, scores them by likelihood and danger, surfaces the hidden assumption everyone treated as settled, and hands you a revised plan with owners, early-warning metrics, and kill criteria.
Here’s what’s inside:
The exact, complete
SKILL.mdfile you can upload to Claude today.The decision science behind why pre-mortems work, in language you can use to win over a skeptical team.
Five worked playbooks for product launches, fundraises, hires, strategy decisions, and investments.
A real example output showing a launch getting torn apart and rebuilt into something that survives.
The “weak input vs. strong input” rule that separates a useless pre-mortem from one that changes a decision.
How it plugs into the 12-skill AI startup operating system as the missing 13th skill - the one that stress-tests every plan the other twelve produce before you execute.
Step-by-step install instructions for both Claude Code and Claude AI.
If you make high-stakes bets, this is the cheapest insurance you’ll ever buy: a few minutes of imagined failure in exchange for the failures you never have to live through.