writing-skills

Enforce RED-GREEN-REFACTOR cycles when authoring AI Skills.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/pearjelly/turbooknow --skill writing-skills-pearjelly
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/pearjelly/turbooknow/tree/main/.agent/skills/writing-skills
Command: npx skills add https://github.com/pearjelly/turbooknow --skill writing-skills-pearjelly

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

This Skill provides a rigorous, test-driven development (TDD) framework for creating and refining AI Skills, ensuring they are effective, discoverable, and robust under pressure.

Core Features & Use Cases

  • TDD for Documentation: Adapts the RED-GREEN-REFACTOR cycle to skill authoring.
  • Pressure Testing: Develops scenarios to test skills under realistic constraints (time, sunk cost, authority).
  • Rationalization Countermeasures: Identifies and explicitly addresses common excuses agents use to bypass rules.
  • Use Case: When developing a new skill for code review, you would first write scenarios where an agent skips review under time pressure (RED), then write the skill to enforce review (GREEN), and finally add counters for rationalizations like "it's just a small change" (REFACTOR).

Quick Start

Use the writing-skills skill to create a new skill document following TDD principles.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
What is pressure testing for agent behavior in skill development?

Rationalization countermeasures identify and explicitly address common excuses agents use to bypass rules. For example, they counter statements like "it's just a small change" by enforcing strict review protocols regardless of the perceived change size.

How do I stop an AI agent from skipping rules under time pressure?

To stop an AI agent from skipping rules under time pressure, you write specific pressure test scenarios during the RED phase of TDD. You then author skill documentation that explicitly enforces compliance and includes rationalization countermeasures for time-based excuses.

Can I use this TDD methodology to edit existing AI skills?

Yes, you can use this TDD methodology to edit existing AI skills. The framework applies to both creating new skills and refining existing ones by verifying their effectiveness through pressure scenarios and adding rationalization countermeasures to ensure robustness.

What are the limitations of using TDD for skill documentation?

Using TDD for skill documentation requires strict adherence to RED-GREEN-REFACTOR cycles and explicit testing against agent failure modes. It demands anticipating rationalizations and pressure scenarios upfront, making the initial authoring phase highly deliberate and time-intensive.