writing-skills

Develop and validate Claude Skills through test-driven RED-GREEN-REFACTOR cycles.

Updated Feb 11, 2016
One-click install
npx skills add https://github.com/Ehrax/dotfiles --skill writing-skills-ehrax
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/Ehrax/dotfiles/tree/main/configs/agents/skills/writing-skills
Command: npx skills add https://github.com/Ehrax/dotfiles --skill writing-skills-ehrax

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Writing Skills provide a disciplined framework for creating and validating Claude Skills through test-driven development, preventing vague guidance and ensuring reusable patterns.

Core Features & Use Cases

  • Structured Skill metadata (YAML frontmatter with name and description) for consistent discovery
  • TDD-driven workflow (RED-GREEN-REFACTOR) to test and refine skills before deployment
  • Progressive disclosure guidance with inline patterns, references, and flowcharts to keep context lean
  • Quality controls including red flags, rationalization tables, and refactoring checks

Quick Start

Follow the RED-GREEN-REFACTOR cycle to create, validate, and bulletproof a Skill before deployment.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I use test-driven development to write Claude skills?

To write Claude skills with TDD, first identify baseline agent failures by running scenarios without the skill and recording exact rationalizations. Next, develop a minimal skill addressing those failures and re-run tests under pressure to verify compliance before delivering a self-contained SKILL.md.

What is the best way to prevent vague guidance when creating documentation skills?

Preventing vague guidance in documentation skills requires a disciplined TDD framework using the RED-GREEN-REFACTOR cycle. This process tests and refines skills iteratively, adding quality controls like red flags and rationalization tables to ensure reusable patterns before deployment.

What metadata structure is required for a self-contained SKILL.md file?

A self-contained SKILL.md file requires YAML frontmatter with name and description fields for consistent discovery. It must also contain a concise description, inline examples, red flags, rationalization tables, and a plan to close loopholes.

How do I close loopholes in AI agent workflow documentation?

Closing loopholes in AI agent workflow documentation involves identifying baseline agent failures by recording exact rationalizations during scenario testing. You then develop a minimal skill addressing those failures and explicitly include a plan to close loopholes within the SKILL.md file.

Does writing Claude skills require progressive disclosure?

Yes, writing Claude skills uses progressive disclosure guidance with inline patterns, references, and flowcharts to keep context lean. This structured approach prevents context overload while delivering precise instructions to the agent.