Test Skill

Before you ship a skill that's supposed to enforce discipline, run it through scenarios where an agent actually wants to break the rules. This applies TDD's red-green-refactor cycle to process documentation: baseline test without the skill to watch agents fail and rationalize, write the skill to address those specific failures, then pressure test with combined stressors like time constraints plus sunk costs plus authority pressure. The rationalization table approach is clever. You capture exact excuses agents use ("keep as reference while writing tests first") and plug each loophole explicitly. Worth the setup cost for any skill with compliance costs attached, like testing requirements or code review checklists.

skills.sh installs: 1.2K
GitHub stars: 1.8K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add neolabhq/context-engineering-kit --skill test-skill --agent claude-code

Installs into .claude/skills of the current project.

Files

Testing Skills With Subagents

Test skill provided by user or developed before.

Overview

Testing skills is just TDD applied to process documentation.

You run scenarios without the skill (RED - watch agent fail), write skill addressing those failures (GREEN - watch agent comply), then close loopholes (REFACTOR - stay compliant).

Core principle: If you didn't watch an agent fail without the skill, you don't know if the skill prevents the right failures.

REQUIRED BACKGROUND: You MUST understand superpowers:test-driven-development before using this skill. That skill defines the fundamental RED-GREEN-REFACTOR cycle. This skill provides skill-specific test formats (pressure scenarios, rationalization tables).

Complete worked example: See examples/CLAUDE_MD_TESTING.md for a full test campaign testing CLAUDE.md documentation variants.