pre-mortem

Pre-mortem risk check

  • Thinking
  • MIT

The agent imagines your launch or plan has already failed and works backwards to explain why. It sorts the risks into real threats, overblown worries and unspoken concerns, then marks each one as fix before launch, fix soon after, or just watch.

When to use it

A week or two before a launch, campaign or big purchase, while there is still time to fix what could sink it.

What it can access

  • Uses the internet
  • Reads or writes files

Before you enable it: Reads the plan file you give it and may search the web about competitors or market conditions. If that is a concern, remove confidential details from the plan first.

Works with

How to install

This skill uses the open Agent Skills format (a folder with a SKILL.md file), so the same file works in several agents.

  • Claude Code: put the folder in ~/.claude/skills/
  • Codex: put the folder in ~/.agents/skills/
  • Other agents that read SKILL.md: follow that agent's skills docs.

Public skill hubs have carried malicious skills. Wize reads every skill before installing it and sets it up so a person approves anything it sends.

Real cases for this kind of work

Leadership & reporting →

Want these running for your team?

Wize installs the skills you need, connects your tools and runs the agents every month. A person approves anything that goes out.

Have Wize install and check it