skills.rest
kylehughes
kylehughesCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

writing-agent-skills

Guide authoring and structuring AI Agent Skills with naming conventions and validation checklists.

3|Updated Jun 12, 2014
Search Tags:#documentation#prompt engineering#ai development#agent skills#skill authoring#skill structure
One-click install
npx skills add https://github.com/kylehughes/knapsack --skill writing-agent-skills
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-agent-skills
Source: https://github.com/kylehughes/knapsack/tree/main/dotfiles/link/claude/skills/writing-agent-skills
Command: npx skills add https://github.com/kylehughes/knapsack --skill writing-agent-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured approach to authoring and refining Agent Skills, ensuring they are discoverable, efficient, and adhere to best practices for AI agent interaction.

Core Features & Use Cases

  • Skill Structure Guidance: Learn how to organize SKILL.md, REFERENCE.md, and scripts for optimal context management.
  • Naming Conventions: Understand the importance of gerund-based naming for discoverability.
  • Description Best Practices: Write effective, third-person descriptions that trigger agent usage.
  • Use Case: When creating a new skill for analyzing code, use this skill to ensure the SKILL.md is concise, the description is clear, and any helper scripts are well-defined.

Quick Start

Use the writing-agent-skills skill to create a new skill for processing user feedback.

Frequently Asked Questions about writing-agent-skills

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

FAQPage Schema
How do I structure AI agent skills for better discoverability and context management?▼

Structure AI agent skills by organizing context progressively across SKILL.md and REFERENCE.md files. This ensures agents efficiently load concise instructions while retaining access to detailed reference data when needed.

What is the best way to name and describe agent skills so AI models trigger them correctly?▼

Name agent skills using gerund-based conventions and write third-person descriptions packed with trigger keywords. This approach maximizes discoverability and ensures the AI agent accurately identifies when to invoke the skill.

How do I create helper scripts for deterministic tasks within an AI agent skill?▼

Create utility scripts for deterministic tasks by defining them clearly within your skill structure. The skill authoring process supports integrating these scripts to handle precise operations, keeping the SKILL.md focused on context.

Does this skill authoring approach work for building code analysis and feedback processing agents?▼

Yes, this skill authoring approach works for building code analysis and feedback processing agents. It provides a validation checklist to ensure helper scripts and SKILL.md files meet efficiency standards for any specific use case.

Why should I use progressive disclosure instead of putting all instructions in one file?▼

Use progressive disclosure to prevent overwhelming the AI agent with excessive context. Separating concise logic in SKILL.md from detailed references in REFERENCE.md optimizes token usage and improves agent reasoning performance.

Related Skills

Explore More Software Engineering Skills→
kk20300113-pngkk20300113-png

writing-skills

Apply test-driven development cycles to author and validate agent skills.

Community
Advanced
PowerBeefPowerBeef
350

qwenvoice-packaged-validation

Validate QwenVoice builds, packaged apps, and release artifacts against the repo harness.

Community
Advanced
Manvendra08Manvendra08
1

token-efficiency

Reduce token consumption and context overflow in LLM interactions.

Community
Basic
Neko-ProtocolNeko-Protocol
6

agent-md-refactor

Refactor monolithic agent instruction files into linked Markdown hierarchies.

Official
Intermediate