testing-skills-with-subagents

Tests documentation skills using a RED-GREEN-REFACTOR workflow with subagents.

1|1|Updated Sep 22, 2025
One-click install
npx skills add https://github.com/SkogAI/skillsandknowledge --skill testing-skills-with-subagents-skogai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: testing-skills-with-subagents
Source: https://github.com/SkogAI/skillsandknowledge/tree/main/skills/claude/skills/testing-skills-with-subagents
Command: npx skills add https://github.com/SkogAI/skillsandknowledge --skill testing-skills-with-subagents-skogai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams verify new or edited skills before deployment by applying a RED-GREEN-REFACTOR cycle to process documentation, forcing baseline failures and iterative improvements.

Core Features & Use Cases

  • RED phase baseline testing without the skill to capture failures and rationalizations verbatim.
  • GREEN phase: implement minimal skill addressing observed failures.
  • REFACTOR phase: close loopholes by documenting new rationalizations and updating rules.
  • Use Case: When validating CLAUDE.md or other skills before production to ensure compliance under pressure.

Quick Start

  • Run a baseline scenario without the skill to observe failures.
  • Then write or adjust the skill and re-run to verify compliance.
  • Use the provided examples/CLAUDE_MD_TESTING.md to model testing campaigns.

Frequently Asked Questions about testing-skills-with-subagents

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

FAQPage Schema
How do I test AI skills under pressure to prevent rationalization?▼

You test skills under pressure by applying a RED-GREEN-REFACTOR workflow across baseline testing, skill implementation, and regression checks using subagents to force baseline failures and curb bias.

What is the RED-GREEN-REFACTOR workflow for skill documentation testing?▼

The RED-GREEN-REFACTOR workflow captures baseline failures without the skill, implements minimal skill adjustments to pass tests, and then closes loopholes by documenting new rationalizations and updating rules.

How do I validate CLAUDE.md or other skills before production deployment?▼

You validate skills before production by running a baseline scenario without the skill to observe failures, then writing or adjusting the skill and re-running to verify compliance under pressure.

Do I need a SKILL.md file to run skill documentation testing?▼

Yes, skill documentation testing requires a SKILL.md file at the root with a YAML frontmatter containing name and description, and it may reference optional scripts, references, and assets directories.

How does baseline testing without the skill capture rationalizations?▼

Baseline testing without the skill captures failures and rationalizations verbatim during the RED phase, establishing a clear record of how the system behaves before any skill-driven constraints are applied.

When should I not use subagent-based skill testing?▼

Subagent-based skill testing is not suitable when there is no SKILL.md file at the root or when the project lacks the required YAML frontmatter containing the name and description fields.