write-skill

Create and improve Claude Code skills with SKILL.md structure and validation checks.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/lzlatkin/dots --skill write-skill-lzlatkin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-skill
Source: https://github.com/lzlatkin/dots/tree/main/agents/skills/write-skill
Command: npx skills add https://github.com/lzlatkin/dots --skill write-skill-lzlatkin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you create or improve Claude Code skills (slash commands) by ensuring they follow best practices for structure, dynamic context, and safety.

Core Features & Use Cases

  • New Skill Creation: Guides you through designing and writing a SKILL.md file from scratch.
  • Skill Improvement: Helps refine existing skills by applying best practices.
  • Use Case: You need to create a new /summarize-meeting skill. You can use this write-skill to generate the initial SKILL.md structure, including context gathering and instruction writing.

Quick Start

Use the write-skill skill to create a new skill that fetches the current git branch and its commits.

Frequently Asked Questions about write-skill

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

FAQPage Schema
How do I create a Claude Code skill with best practices for dynamic context?

To create a Claude Code skill, define the frontmatter, design context sections using shell commands, and write instructions. This process ensures dynamic context injection and safety protocols while validating against common pitfalls like improper `$()` usage.

What is the best way to structure a Claude Code slash command?

The best way to structure a slash command is to follow a standardized `SKILL.md` format. This involves defining frontmatter, creating context sections with shell commands, and writing clear instructions while applying validation checks for common errors.

Can I use shell commands to inject dynamic context into a Claude Code skill?

Yes, you can use shell commands to inject dynamic context into a Claude Code skill. The skill guides you through designing context sections that execute shell commands to gather real-time data for the prompt.

What common pitfalls should I avoid when writing Claude Code skills?

When writing Claude Code skills, avoid pitfalls like improper `$()` usage and incorrect error handling. The skill creation process includes validation checks to ensure your skill adheres to safety protocols and structural best practices.

Do I need to write a SKILL.md file from scratch to automate a task like fetching git commits?

No, you do not need to write a `SKILL.md` file entirely from scratch. You can use a skill creation tool to generate the initial structure, including context gathering for git branches and commits, while ensuring adherence to best practices.