skills.rest
raas-dev
raas-devOFFICIAL
Explore all skills by this author →
GitHub

writing-skills

Apply Test-Driven Development to create and refine Claude Skills.

19|4|Updated Dec 23, 2021
Search Tags:#skill development#documentation#TDD#AI agents#agent behavior#skill optimization#rationalization
One-click install
npx skills add https://github.com/raas-dev/configent --skill writing-skills
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/raas-dev/configent/tree/main/etc/claude-code/skills/writing-skills
Command: npx skills add https://github.com/raas-dev/configent --skill writing-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill applies Test-Driven Development (TDD) to the creation and editing of Claude Skills, ensuring they are robust, resist rationalization, and effectively guide agents under pressure, preventing agents from bypassing critical rules.

Core Features & Use Cases

  • TDD for Documentation: Guides through the RED-GREEN-REFACTOR cycle for skill writing, using subagents for testing compliance and effectiveness.
  • CSO (Claude Search Optimization): Provides guidelines for naming, descriptions, and keyword usage to optimize skill discovery by future Claude instances.
  • Bulletproofing Against Rationalization: Teaches how to explicitly close loopholes and counter common agent excuses, ensuring disciplined adherence to skill rules.
  • Skill Structure & Content Guidelines: Offers best practices for organizing SKILL.md, using flowcharts, code examples, and managing file organization for clarity.
  • Use Case: When developing a new skill, use this to ensure it's thoroughly tested against agent behavior, clearly discoverable, and effectively enforces desired workflows, making your skills truly impactful.

Quick Start

TDD Mapping for Skills

Test case: Pressure scenario with subagent

Production code: Skill document (SKILL.md)

Test fails (RED): Agent violates rule without skill (baseline)

Test passes (GREEN): Agent complies with skill present

Refactor: Close loopholes while maintaining compliance

Write test first: Run baseline scenario BEFORE writing skill