Agent skill

create-skill

Create a new agent skill following the Agent Skills specification. Use this when the user asks to create a new skill or update an existing one to match the standard.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/create-skill-hhk7734-hhk7734-2

SKILL.md

Create Skill

When to use this skill

Use this skill when the user wants to create a new agent skill or explicitly asks to "create a skill". This skill provides the standard procedure and specification for creating high-quality, compliant agent skills.

Prerequisite: Understand the Goal

  1. Identify the Skill Name:
    • must be 1-64 characters.
    • only lowercase alphanumeric (a-z, 0-9) and hyphens (-).
    • cannot start or end with a hyphen.
    • cannot have consecutive hyphens (--).
  2. Identify the Purpose: Understand what task the skill is solving to write a good description.

Step 1: Create Directory Structure

Create a new directory with the exact name of the skill.

<skill-name>/
  ├── SKILL.md  (Required)
  ├── scripts/  (Optional: executable code)
  ├── references/ (Optional: docs, forms)
  └── assets/   (Optional: templates, images)

Step 2: Create SKILL.md

The SKILL.md file is the core of the skill. It MUST strictly follow this format:

Frontmatter

The file must start with a YAML frontmatter block:

yaml
---
name: <skill-name>
description: <text description of what the skill does and when to use it>
# Optional fields:
# license: <license type>
# compatibility: <system requirements>
# metadata:
#   author: <name>
#   version: "1.0"
---

Critical Rules:

  • name in frontmatter MUST match the directory name.
  • description should include keywords to help agents find the skill (1-1024 chars).

Body

The body should be standard Markdown containing:

  • Header: # Title of Capability
  • When to use: A section explaining the specific scenarios where this skill is applicable.
  • Instructions: Step-by-step procedures the agent should follow.
    • Be prescriptive and clear.
    • Reference scripts or assets using relative paths (e.g., scripts/my-script.py).

Step 3: Populate Optional Directories (If needed)

  • scripts/: For code that needs to be executed (e.g., Python scripts). Ensure they are self-contained.
  • references/: For static knowledge, cheat sheets, or lookup tables.
  • assets/: For templates the agent should copy or use.

Validation

After creating the files, ensure:

  1. The directory name and name field match exactly.
  2. The YAML frontmatter is valid.
  3. The description clearly explains when to use the skill.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results