skill-writing-guide

Provides guidelines and instructions for creating Agent Skills in the open standard format.

2|1|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/vosslab/vosslab-skills --skill skill-writing-guide-vosslab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-writing-guide
Source: https://github.com/vosslab/vosslab-skills/tree/main/skills/skill-writing-guide
Command: npx skills add https://github.com/vosslab/vosslab-skills --skill skill-writing-guide-vosslab

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides a comprehensive guide to creating and improving Agent Skills using the open standard format, covering metadata, directory structure, writing instructions, and best practices.

Core Features & Use Cases

  • Format Guidance: Offers the open standard format details, required frontmatter, and Markdown body structure.
  • Directory Layout: Explains the minimum and full directory structures for skills.
  • Description Writing: Includes rules for writing effective descriptions to trigger the skill correctly.
  • Progressive Disclosure: Describes how skills manage context efficiently by loading metadata, instructions, and resources.
  • Resource Usage: Details when and how to use scripts, references, and assets within the skill.
  • UI Metadata: Provides guidance on how to structure agents/openai.yaml for UI display on skill lists.
  • Repo-Specific Conventions: Outlines conventions followed in this repository for skills development.
  • Skill Naming: Defines rules for naming skills in the repository.
  • What Not to Include: Lists what should not be included in a skill folder.
  • Checklist: Provides a detailed checklist for verifying skill completeness before shipping.
  • Further Reading: Links to additional resources and specifications.
  • Delegated Execution: Explains the philosophy behind assigning subagents for skill execution.

Quick Start

Use the skill-writing-guide skill to create or enhance your own Agent Skill.

Frequently Asked Questions about skill-writing-guide

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

FAQPage Schema
How do I create an Agent Skill using the open standard format?

Agent Skills manage context efficiently by progressively loading metadata, instructions, and resources only when needed. This progressive disclosure mechanism ensures the agent processes relevant information without exceeding context limits.

What directory structure is required for an Agent Skill?

An Agent Skill directory structure requires a SKILL.md file at minimum, with optional folders for scripts, references, assets, and agents to organize resources and UI metadata like openai.yaml.

How does progressive disclosure work for agent skills?

Agent skills manage context efficiently by progressively loading metadata, instructions, and resources only when needed, preventing context overflow and optimizing execution.

What should not be included in an agent skill folder?

An agent skill folder should not include files or modules that violate repository-specific conventions. Always verify the provided checklist to ensure excluded items are removed before shipping.

Can I use scripts and references within an agent skill?

Yes, you can use scripts and references within an agent skill to handle resource usage. The open standard format provides detailed guidance on when and how to incorporate these components effectively.

Do I need subagents for agent skill execution?

Delegated execution using subagents is a supported philosophy for skill execution. This approach assigns subagents to handle tasks, optimizing the skill's processing logic and overall performance.