- Home
- Skills
- Ntcoding
- Claude Skillz
- Challenge That
challenge-that_skill
- Python
247
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill ntcoding/claude-skillz --skill challenge-that- SKILL.md4.3 KB
Overview
This skill forces critical evaluation of proposals, requirements, or decisions by attacking them from five adversarial perspectives. It triggers when consensus is premature or assumptions go unchallenged and returns concrete, actionable concerns to guide decision-making. Invoke with /challenge-that to force a stronger review before proceeding.
How this skill works
When invoked, the skill identifies the accepted or proposed item, states it clearly, and then analyzes it from five distinct perspectives: Skeptic, Pragmatist, Edge Case Hunter, Structural Critic, and Root Cause Analyst. Each perspective raises specific, testable concerns and the skill synthesizes those into a short list of prioritized issues and next steps.
When to use it
- A proposal is accepted with comments like "sounds good" or "let's go with"
- Design decisions are made quickly with unstated tradeoffs
- Requirements feel under-examined or assumptions are unchallenged
- You want to stress-test a plan before committing resources
- Something seems off but you can't articulate why
Best practices
- Attach the minimal context: the accepted statement and any supporting claims
- Be specific about scope so challenges target the right decision
- Use the output as a checklist of investigations, not final judgments
- Prioritize concerns by impact and ease of mitigation
- Address the highest-impact concerns before committing resources
Example use cases
- A PM says "deploy this feature next week" after a brief demo; run /challenge-that to surface hidden risks
- A dev proposes a quick patch that changes architecture; use the skill to expose maintainability and failure-mode risks
- Acceptance of a vendor recommendation without validation; challenge evidence and cost assumptions
- Requirements written hastily; challenge to reveal ambiguous or missing acceptance criteria
- Team reaches consensus after a short discussion; force adversarial review to avoid groupthink
FAQ
No. It focuses on ideas and decisions, not people, and frames challenges as actionable investigations.
What if a perspective finds nothing wrong?
The skill will say so explicitly, but it still synthesizes overall confidence and any lower-priority checks to validate that view.