writing-great-skills

Define principles for designing, editing, and maintaining predictable AI agent skills.

Updated Apr 8, 2026

Search Tags:#architecture #refactoring #prompt-engineering #ai-agents #skill-design #predictability

Ready to run

npx skills add https://github.com/tonomb/personal-pnl --skill writing-great-skills-tonomb

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: writing-great-skills
Source: https://github.com/tonomb/personal-pnl/tree/main/.agents/skills/writing-great-skills
Command: npx skills add https://github.com/tonomb/personal-pnl --skill writing-great-skills-tonomb

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill addresses the challenge of creating AI-agent behaviors that are inconsistent or unpredictable, providing a rigorous framework to ensure agents follow the same process every time they are invoked.

Core Features & Use Cases

Predictability Framework: Defines the levers of invocation, information hierarchy, and steering to ensure agent reliability.
Pruning Discipline: Offers clear methodologies to identify and remove sprawl, no-ops, and duplication to keep skills lean and efficient.
Use Case: Use this skill to audit your existing agent library, refactor overly verbose instructions into compact leading words, and optimize your skill architecture for lower context load.

Quick Start

Invoke the writing-great-skills skill to analyze my current prompt structure and suggest improvements for better predictability.

Frequently Asked Questions about writing-great-skills

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

FAQPage Schema

How do I make my AI agent skills more predictable and consistent?▼

To make AI agent skills predictable, apply a framework of structured information hierarchies, progressive disclosure, and rigorous pruning of redundant instructions to ensure consistent behavior across every invocation.

What is the best way to refactor verbose AI prompts for lower context load?▼

The best way to refactor verbose AI prompts is by applying pruning discipline to identify and remove sprawl, no-ops, and duplication, transforming them into compact leading words for lower context load.

Why does my AI agent behave inconsistently even with detailed instructions?▼

AI agents behave inconsistently when instructions lack a defined information hierarchy and steering levers. Establishing a structured predictability framework ensures the agent follows the exact same process every time.

How do I audit my existing AI agent library for skill design efficiency?▼

Audit your AI agent library by analyzing current prompt structures against predictability principles, identifying sprawl and duplication, and refactoring instructions to optimize the overall skill architecture.

When should I use progressive disclosure in AI prompt engineering?▼

Use progressive disclosure in AI prompt engineering when designing modular, agent-discoverable skill units to manage context load and ensure predictable agent behavior within a larger system.