refute

Execute adversarial tests to verify software completion claims.

Updated Jul 10, 2026
One-click install
npx skills add https://github.com/ronniepinnell/casper --skill refute
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: refute
Source: https://github.com/ronniepinnell/casper/tree/main/skills/refute
Command: npx skills add https://github.com/ronniepinnell/casper --skill refute

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps prevent false claims of completion by adversarially testing them before they are accepted.

Core Features & Use Cases

  • Adversarial Claim Verification: Before accepting any "it works / it's done / it's fixed" claim, construct the concrete input that would break it and run it.
  • Before Commits, PRs, Issues: Use to verify claims before committing code, creating pull requests, or closing issues.
  • Use Case: When a developer claims a bug is fixed, use this Skill to test the fix by providing inputs that should fail if the fix is not effective.

Quick Start

Run the /refute command followed by the claim to be verified, e.g., /refute the login fix works.

Frequently Asked Questions about refute

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify a bug fix before committing code?

Adversarial testing validates bug fixes by constructing concrete inputs designed to break the claim. By executing scripts and commands against the fix, you confirm whether the completion claim holds true before committing code.

What is adversarial testing in software validation?

Adversarial testing in software validation is the practice of constructing concrete inputs designed to break a claimed fix. It prevents false claims of completion by actively trying to invalidate the software before accepting it as done.

When should I use claim verification in a development workflow?

You should use claim verification in a development workflow before commits, pull requests, or closing issues. It ensures that any claim stating a feature is implemented or a bug is fixed is actively tested against breaking inputs.

Do I need specific scripts to run adversarial tests on completion claims?

Yes, verifying completion claims requires the ability to execute scripts and commands. The testing process depends on running these scripts to provide inputs that validate whether the claimed bug fix or feature implementation actually works.