writing-skills

Enforce test-driven development for documentation-based Skills via RED-GREEN-REFACTOR loops.

8|1|Updated Mar 1, 2026
One-click install
npx skills add https://github.com/JoeCP17/LLM-Dot-files --skill writing-skills-joecp17
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/JoeCP17/LLM-Dot-files/tree/main/claude/skills/superpowers/writing-skills
Command: npx skills add https://github.com/JoeCP17/LLM-Dot-files --skill writing-skills-joecp17

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents Claude agents from treating skill documentation as optional or interpretive guidance, and instead forces reliable compliance by using a RED-GREEN-REFACTOR loop specifically for process docs.

Core Features & Use Cases

  • TDD for skills: Create pressure scenarios that demonstrate baseline violations, then author the minimum documentation needed for compliance.
  • Rationalization bulletproofing: Identify the exact excuses agents use under pressure and add explicit negations, counter-rules, and red flags to close loopholes.
  • Search/discovery optimized structure: Enforces YAML frontmatter conventions (name, description), strong “Use when...” trigger descriptions, and keyword coverage to improve skill discovery.

Quick Start

Use this skill when writing, editing, or validating a documentation-based Skill before deployment, so you can run pressure scenarios that would otherwise cause agents to rationalize around the rules.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I enforce test-driven development for writing skill documentation?

Test-driven development for skill documentation is enforced by defining pressure-scenario tests that demonstrate baseline violations, then authoring the minimum documentation needed to achieve compliance. A RED-GREEN-REFACTOR loop validates that agents follow instructions.

Why do AI agents rationalize skipping required steps in process documentation?

Agents rationalize skipping required steps in process documentation by treating instructions as non-binding guidance under pressure. This behavior is countered by identifying specific excuses and adding explicit negations, counter-rules, and red flags to close loopholes.

What is a pressure scenario in skill authoring and verification?

A pressure scenario in skill authoring is a test case designed to demonstrate baseline violations where agents are likely to rationalize skipping steps or treating instructions as optional. It validates that documentation closes loopholes via re-testing.

How do I optimize skill discovery using YAML frontmatter?

Optimizing skill discovery using YAML frontmatter requires enforcing specific conventions for the `name` and `description` fields, alongside strong third-person "Use when..." trigger descriptions and keyword coverage to improve search visibility.

Can I use a RED-GREEN-REFACTOR loop for prompt compliance testing?

Yes, you can use a RED-GREEN-REFACTOR loop for prompt compliance testing to verify documentation-based skills. It identifies baseline violations, authors minimum documentation for compliance, and closes loopholes through rationalization bulletproofing and re-testing.

What is the best way to prevent agents from treating documentation as optional guidance?

Preventing agents from treating documentation as optional guidance requires defining pressure scenarios that demonstrate baseline violations, adding explicit loophole-closing counters, and validating compliance through re-testing within a RED-GREEN-REFACTOR loop.