What Could Go Wrong

What Could Go Wrong is an AI Engineering skill for Claude Code, published by kharmanskyi in open-steps.

211 stars55 forkson kharmanskyi/open-stepsAdded 2026/09/05+56% in starsRepository updated 2026/09/03

agent-skillagent-skillsai-agentsclaude-codeclaude-code-pluginclaude-code-skillsclaude-skillprompt-optimizationvibe-coding

Install in seconds

Install What Could Go Wrong

Copy What Could Go Wrong into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/kharmanskyi/open-steps/tree/main/skills/os-what-could-go-wrong ~/.claude/skills/os-what-could-go-wrong

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/kharmanskyi/open-steps.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: skills/os-what-could-go-wrong/SKILL.md in kharmanskyi/open-steps
Installs to: ~/.claude/skills/os-what-could-go-wrong
Collection: One of 7 skills cataloged from this repository
Category: AI Engineering — 3101 skills

What What Could Go Wrong does

What Could Go Wrong runs a premortem before hard-to-undo decisions, asking what might fail, what’s missing, and why the plan could break. It is used for contracts, launches, migrations, and other one-way changes.

What Could Go Wrong is cataloged under AI Engineering on DirSkills. What Could Go Wrong comes from a repository tagged agent-skill, agent-skills, ai-agents, claude-code and claude-code-plugin.

Documentation

README

Excerpt

os-what-could-go-wrong

Assume it already failed, then work backwards to find out why - while there is still time to change it. The attack runs in a fresh agent that had no part in the decision, because an agent that helped shape one reviews it far too gently: it defends its own reasoning, and it misses the thing that kills the plan out of politeness. This skill writes the brief, hands it over, and passes the answer back without softening it.

os-ask-simple screens a choice before the user picks one. This one runs after the choice is made and before it can no longer be taken back.

Language

This is the opening of the README. Read the full README on GitHub.

Frequently asked about What Could Go Wrong

What else does kharmanskyi publish alongside What Could Go Wrong?
What Could Go Wrong is one of 7 skills that DirSkills catalogs from kharmanskyi/open-steps, the repository it ships in. Its siblings there include Ask Simple, Check Work and Done Or Not. Each one is a separate skill with its own page in this directory, installs the same way What Could Go Wrong does, and is maintained by kharmanskyi in that same repository. The rest of the collection is listed on the kharmanskyi/open-steps page.
How does What Could Go Wrong compare to other AI Engineering skills?
What Could Go Wrong ranks #2682 by stars among the 3101 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of What Could Go Wrong against them. Open each page to compare what they document and how they install.