skills.rest
ninjasitm
ninjasitmCOMMUNITY
Explore all skills by this author →
GitHub

writing-skills

Design and test Skills using a RED-GREEN-REFACTOR cycle.

2|1|Updated Dec 9, 2025
Search Tags:#workflow#testing#tdd#documentation#rationalization#skill-writing
One-click install
npx skills add https://github.com/ninjasitm/ai-assisted-dev-toolkit --skill writing-skills-ninjasitm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/ninjasitm/ai-assisted-dev-toolkit/tree/main/src/monorepo/.agents/skills/writing-skills
Command: npx skills add https://github.com/ninjasitm/ai-assisted-dev-toolkit --skill writing-skills-ninjasitm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Creating and validating Skills is error-prone without a disciplined process. This guide provides a TDD-inspired framework to design, test, and harden Skills before deployment.

Core Features & Use Cases

  • Red-Green-Refactor workflow for skill creation, including baseline tests, minimal Skill, and explicit counters to prevent regressions.
  • Rationalization tracking with red flags and a structured rationalization table to surface and close loopholes.
  • Deployment readiness requiring a testing checklist and progressive-disclosure approach to keep Skill bodies concise.

Quick Start

Start by writing a baseline failing test (RED) against a task without a Skill, then implement the minimal Skill (GREEN) to satisfy the test, and finally refactor to close loopholes.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I test and validate agent skill instructions before deployment?▼

A TDD-inspired workflow validates skill creation by enforcing baseline failure tests, implementing minimal instructions to pass them, and tracking rationalizations to close loopholes before deployment.

What is the RED-GREEN-REFACTOR cycle for creating agent skills?▼

This TDD-inspired workflow requires writing a baseline failing test without a Skill, implementing the minimal Skill to pass it, and refactoring instructions with explicit counters to prevent regressions across projects.

How do I prevent regressions when editing existing skills?▼

Use explicit counters and a structured rationalization tracking table to surface red flags, close loopholes, and prevent regressions when editing or validating instructions across agent projects.

What is rationalization tracking in skill documentation?▼

Rationalization tracking uses a structured table to surface red flags and close loopholes in skill instructions, ensuring robust and reusable documentation across different agent deployment contexts.

Can I use this testing workflow for skills in any agent context?▼

Yes, this TDD-inspired testing workflow applies when creating, editing, or validating skills before deployment in any agent context, ensuring robust and reusable instructions across diverse project environments.

What deployment checks are required before hardening a skill?▼

Deployment checks require completing a testing checklist and applying a progressive-disclosure approach to keep Skill bodies concise, ensuring instructions are fully hardened and ready for any agent context.

Related Skills

Explore More Content & Communication Skills→
pv-udpvpv-udpv
4

Testing Skills With Subagents

Run RED-GREEN-REFACTOR tests to harden skill documentation and close loopholes.

Community
Advanced
JonghakseoJonghakseo
13

skill-creator

Create, modify, and validate Pi Agent Skills with metadata and triggers.

Community
Advanced
h4v1dh4v1d

testing-skills-with-subagents

Apply RED-GREEN-REFACTOR testing to process documentation and capture rationalizations verbatim.

Community
Advanced
WcofWcof
1

pm-skillauthor

Create agent skills via RED-GREEN-REFACTOR with SKILL.md and pressure tests.

Community
Advanced