creating-skills

Document techniques using a RED-GREEN-REFACTOR test-driven development cycle.

Updated Jan 14, 2024
One-click install
npx skills add https://github.com/michellepellon/dotfiles --skill creating-skills-michellepellon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: creating-skills
Source: https://github.com/michellepellon/dotfiles/tree/main/.claude/skills/creating-skills
Command: npx skills add https://github.com/michellepellon/dotfiles --skill creating-skills-michellepellon

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the problem of creating unreliable, poorly documented processes. It ensures that every new skill or documented technique is robust, clear, and effective by applying a rigorous Test-Driven Development (TDD) approach to documentation. This prevents future AI agents from misinterpreting or failing to apply the documented knowledge, saving significant time and rework.

Core Features & Use Cases

  • TDD for Documentation: Guides you through a RED-GREEN-REFACTOR cycle for process documentation, ensuring clarity and correctness.
  • Bulletproof Skills: Helps identify and close loopholes in instructions, making skills resilient to agent rationalizations and pressures.
  • Use Case: When you've developed a new technique for debugging a specific type of error, use this skill to document it. You'll first test how an agent fails without the documentation, then write the skill to address those failures, and finally refactor until the agent consistently applies the technique correctly under pressure.

Quick Start

Announce at start:

"I'm using the creating-skills skill to build and validate this skill."

RED Phase - Write Failing Test:

Create pressure scenarios (3+ combined pressures for discipline skills)

Run scenarios WITHOUT skill - document baseline behavior verbatim

Identify patterns in rationalizations/failures

GREEN Phase - Write Minimal Skill:

Name describes what you DO or core insight

YAML frontmatter with rich when_to_use (include symptoms!)

Keywords throughout for search (errors, symptoms, tools)

Clear overview with core principle

Address specific baseline failures identified in RED

Code inline OR @link to separate file

One excellent example (not multi-language)

Run scenarios WITH skill - verify agents now comply