writing-skills

Apply TDD with RED-GREEN-REFACTOR cycles to SKILL.md documentation.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enforces a Test-Driven Development approach to crafting Skill documentation, ensuring that every Skill begins with a failing baseline, is implemented minimally, and is refactored to close loopholes before deployment.

Core Features & Use Cases

  • RED-GREEN-REFACTOR cycles; Frontmatter validation; Subagent-based testing; Clear deployment checklist; Progressive disclosure; One SKILL.md per Skill.
  • Structured SKILL.md design: frontmatter with name and description, body with guidance.
  • Evidence-based iteration: documents rationalizations and test results to reduce drift.

Quick Start

Draft SKILL.md for a new Skill with frontmatter and a concise description. Then write a pressure scenario to baseline behavior without the Skill, implement the Skill, and re-test until all baselines pass. Finally deploy and monitor.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I apply TDD to writing documentation for AI agents?

Apply TDD to process documentation by establishing a failing baseline scenario, implementing the skill minimally to pass that test, and refactoring to close loopholes before deployment. This enforces a RED-GREEN-REFACTOR cycle for bulletproof agent documentation.

What is a RED-GREEN-REFACTOR workflow for skill documentation?

A RED-GREEN-REFACTOR workflow for skill documentation means writing a pressure scenario that fails without the skill, implementing the skill minimally to make it pass, and refactoring the documentation to eliminate rationalizations and close loopholes before deployment.

How do I verify process documentation before deploying it to Claude-like agent workflows?

Verify process documentation by documenting testing with subagents. You write a pressure scenario to baseline behavior without the skill, implement the skill, and re-test with subagents until all baselines pass before deploying to Claude-like agent workflows.

What's the best way to structure a SKILL.md file for AI agents?

Structure a SKILL.md file by including frontmatter with the name and description, followed by a body with clear guidance. Maintain one SKILL.md per skill and use progressive disclosure to ensure agents receive validated, structured instructions.

Does skill-writing require subagent testing for frontmatter validation?

Yes, skill-writing uses subagent testing to verify frontmatter validation and overall effectiveness. Testing with subagents ensures the frontmatter contains the required name and description, and that the skill behaves correctly before deployment.

Why does my process documentation drift from the intended agent behavior?

Process documentation drifts when rationalizations and test results are not explicitly recorded during iteration. Using an evidence-based TDD approach reduces drift by documenting test outcomes and closing loopholes before the skill is deployed.