What are Claude Code Skills?

Claude Code Skills are reusable prompt templates stored as markdown files in your project's .claude/skills/ directory. They let you codify best practices and common workflows into slash commands that Claude follows consistently.

How do I create a Claude Code Skill?

Create a markdown file in .claude/skills/ with instructions for Claude. The filename becomes the command name — for example, .claude/skills/code-review.md can be invoked with /code-review during a Claude Code session.

Can I share skills with my team?

Yes. Skills stored in .claude/skills/ can be committed to version control so every team member has access to the same standardized workflows. You can also organize skills into subdirectories by category.

What is the difference between project and global skills?

Project skills (.claude/skills/) are specific to the current project and are committed to git. Global skills (~/.claude/skills/) are personal skills available across all your projects.

SKILL_SLUG

New

GitHub TrendingGeneralby Gamma-Software

ONE sentence — what the skill does AND when to trigger it. Pack trigger phrases a user would actually type ("do X", "fix Y", "set up Z") right into this line; the agent matches on it. Keep it specific and verb-led. Replace this whole block.

Community PluginView Source

Overview

Skill Title

One-paragraph orientation: what this skill installs/does and the shape of the work. Lead with the outcome, not the mechanism.

code

Step1 → Step2 → [human-in-the-loop gate] → Step3 → Verify → Clean up

1. <First phase>

Concrete, ordered instructions the agent follows. Prefer imperative voice ("Copy X", "Run Y", "Read one generated file and spot-check"). State the why only where a future maintainer would otherwise simplify away something load-bearing.

2. <Second phase>

•Bullet the decision points (single repo vs. workspace, language coverage…).
•Reference bundled assets by path: assets/<file>, scripts/<file>.
•Call out the human-in-the-loop gates explicitly if any.

3. Validate

Don't declare success blind. List the gates: syntax, a real run, an edge-case check, and a measured recall/behavior check. A gate exists because skipping it shipped a bug at least once — say so.

4. Report

Summarize for the user: what was installed/changed, measurements, a gate/pass/evidence table, and what was skipped.

<!-- TEMPLATE NOTES (delete this block when authoring):

•name MUST be the kebab-case slug and MUST match the repo skill dir name.
•description is the single most important line — it drives triggering.

Write it last, after the body exists. Front-load trigger phrases.

•Keep SKILL.md lean; push long material into references/*.md, code into

scripts/ or assets/. The agent reads references on demand.

•Run the skill-creator skill to lint description quality if available.

-->

Install & Usage

Create the skills directory

mkdir -p .claude/skills

Download the skill file

mkdir -p .claude/skills && curl -o .claude/skills/skill-slug.md https://raw.githubusercontent.com/Gamma-Software/skill-template/main/SKILL.md

Invoke in Claude Code

/skill-slug

View source on GitHub

agent

Frequently Asked Questions

What is SKILL_SLUG?

How to install SKILL_SLUG?

To install SKILL_SLUG, create the .claude/skills directory in your project, then run the curl command to download the skill file. Once installed, invoke it in Claude Code with /skill-slug.

What is SKILL_SLUG best for?

SKILL_SLUG is a community categorized under General. It is designed for: agent. Created by Gamma-Software.