writing-great-skills

Clarify skill design vocabulary and structural rules for predictable agent behavior.

16|2|Updated Mar 29, 2018
One-click install
npx skills add https://github.com/khoi/dotfiles --skill writing-great-skills
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-great-skills
Source: https://github.com/khoi/dotfiles/tree/main/chezmoi/dot_agents/skills/writing-great-skills
Command: npx skills add https://github.com/khoi/dotfiles --skill writing-great-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Clarifies the vocabulary and structural rules used to design predictable skills, reducing ambiguity and alignment friction across teams.

Core Features & Use Cases

  • Establishes a shared glossary of terms (e.g., model-invoked vs user-invoked, context load, leading words) to unify skill design decisions.
  • Defines the Information Hierarchy, co-location, and progressive disclosure to keep SKILL.md readable and actionable.
  • Provides best practices for organizing steps, references, branches, and avoiding duplication and no-ops across skills.

Quick Start

Align your skill by mapping its triggers, structure, and references to the glossary and hierarchy.

Frequently Asked Questions about writing-great-skills

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

FAQPage Schema
What is predictable skill design and why does it matter for consistent agent behavior?

Predictable skill design uses a defined vocabulary and structural rules to reduce ambiguity and alignment friction, ensuring consistent agent behavior across both model-invoked and user-invoked skills. It establishes shared terms and an information hierarchy for actionable outputs.

How do I structure a SKILL.md file to maintain a clear information hierarchy?

Structure your SKILL.md using an information hierarchy that co-locates steps, branches, and context pointers. Apply progressive disclosure by placing references behind expandable sections to keep the file readable and actionable without overwhelming the agent's context load.

What are leading words and how do they affect instruction structure in skill design?

Leading words are part of a shared glossary that unifies skill design decisions by establishing clear triggers and branches. They guide the instruction structure to ensure predictable agent behavior and prevent duplicate operations across model-invoked and user-invoked skills.

Does this approach to skill design work for both user-invoked and model-invoked skills?

Yes, the structural rules and glossary apply across both model-invoked and user-invoked skills. It guides description writing, branch logic, and reference placement behind progressive disclosure to ensure consistent behavior regardless of invocation type.

How do I avoid duplication and no-ops when organizing steps and branches in skills?

Avoid duplication and no-ops by following best practices for organizing steps and references within the defined information hierarchy. Map triggers and branches to the shared glossary to ensure each step performs a distinct, actionable operation.