skills.rest
joe-butler-23
joe-butler-23COMMUNITY
Explore all skills by this author →
GitHub

Writing Skills

Create and test AI agent skill documentation using a RED-GREEN-REFACTOR cycle.

Updated Aug 23, 2026
Search Tags:#workflow#testing#quality assurance#documentation#TDD#AI agent#skill creation
One-click install
npx skills add https://github.com/joe-butler-23/nix-config --skill writing-skills-joe-butler-23
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Writing Skills
Source: https://github.com/joe-butler-23/nix-config/tree/main/modules/home/mcp/skills/meta/writing-skills
Command: npx skills add https://github.com/joe-butler-23/nix-config --skill writing-skills-joe-butler-23

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the challenge of creating clear, effective, and robust documentation for AI agents. It prevents the deployment of unclear or untested instructions that lead to agent confusion, errors, and rework, ensuring that skills are bulletproof against rationalization and misinterpretation.

Core Features & Use Cases

  • Test-Driven Documentation (TDD): Applies the RED-GREEN-REFACTOR cycle to skill creation, ensuring documentation is validated against agent behavior.
  • Rationalization Bulletproofing: Provides techniques to close loopholes and prevent agents from finding excuses to deviate from instructions.
  • Skill Structure & Optimization: Guides on organizing SKILL.md content, optimizing for Claude Search Optimization (CSO), and managing token efficiency.
  • Use Case: When developing a new skill, use this guide to ensure it's thoroughly tested with subagents, preventing ambiguity and ensuring consistent, desired behavior from the AI.

Quick Start

To create a new skill, first, define pressure scenarios for a subagent to test without the skill. Document the subagent's failures and rationalizations. Then, write the skill to address these specific issues, and re-test until the subagent complies.