/skill-creator

The /skill-creator skill teaches you how to create effective Skills that Claude can discover and use successfully.

What are Agent Skills?

Agent Skills are modular, filesystem-based capabilities that provide domain-specific expertise through progressive disclosure. They are organized prompts that get loaded on-demand.

Quick Start Workflow

Identify the reusable pattern: What context would be useful for similar future tasks?
Create directory and SKILL.md:
- Directory name: Follow verb-noun convention: create-*, manage-*, setup-*
- YAML frontmatter with name and description
Write concise instructions: Assume Claude is smart. Only add context Claude doesn't have.
Test with real usage: Iterate based on observations.

Example Skill

YAML

---
name: process-pdfs
description: Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files.
---

<objective>
Extract text and tables from PDF files using Python libraries.
</objective>

<quick_start>
Extract text with pdfplumber:

import pdfplumber
with pdfplumber.open("file.pdf") as pdf:
    text = pdf.pages[0].extract_text()
</quick_start>

<advanced_features>
**Form filling**: See [forms.md](forms.md)
</advanced_features>

Required XML Tags

All skills must have:

Tag	Purpose
`<objective>`	What the skill does and why it matters
`<quick_start>`	Immediate, actionable guidance
`<success_criteria>`	How to know it worked

Conditional Tags

Add based on skill complexity:

Tag	When to Use
`<context>`	Background information
`<workflow>`	Step-by-step procedures
`<advanced_features>`	Deep-dive topics
`<validation>`	How to verify outputs
`<examples>`	Multi-shot learning
`<anti_patterns>`	Common mistakes to avoid
`<security_checklist>`	Security patterns

Skill Directory Structure

skills/{skill-name}/
├── SKILL.md              # Main entry point
├── references/           # Additional documentation
│   └── api-reference.md
└── templates/            # Optional templates
    └── example.md

Best Practices

Keep SKILL.md focused - Link to references for details
Use progressive disclosure - Basic info first, advanced later
Include examples - Show expected input/output
Define success criteria - Clear completion indicators

Slash Command Wrapper

After creating a skill, create a lightweight slash command wrapper to make it invocable via /skill-name:

Location: ~/.claude/commands/{skill-name}.md

Template:

YAML

---
description: {Brief description of what the skill does}
argument-hint: [{argument description}]
allowed-tools: Skill({skill-name})
---

<objective>
Delegate {task} to the {skill-name} skill for: #$ARGUMENTS
</objective>

<process>
1. Use Skill tool to invoke {skill-name} skill
2. Pass user's request: #$ARGUMENTS
3. Let skill handle workflow
</process>

<success_criteria>
- Skill successfully invoked
- Arguments passed correctly to skill
</success_criteria>

The slash command's only job is routing—all expertise lives in the skill.

Quick Start Workflow

Identify the reusable pattern: What context would be useful for similar future tasks?

Create directory and SKILL.md:

Directory name: Follow verb-noun convention: create-*, manage-*, setup-*
YAML frontmatter with name and description

Write concise instructions: Assume Claude is smart. Only add context Claude doesn't have.

Test with real usage: Iterate based on observations.

Example Skill

YAML

---
name: process-pdfs
description: Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files.
---

<objective>
Extract text and tables from PDF files using Python libraries.
</objective>

<quick_start>
Extract text with pdfplumber:

import pdfplumber
with pdfplumber.open("file.pdf") as pdf:
    text = pdf.pages[0].extract_text()
</quick_start>

<advanced_features>
**Form filling**: See [forms.md](forms.md)
</advanced_features>

Tag

Purpose

<objective>

What the skill does and why it matters

<quick_start>

Immediate, actionable guidance

<success_criteria>

How to know it worked

Tag

When to Use

<context>

Background information

<workflow>

Step-by-step procedures

<advanced_features>

Deep-dive topics

<validation>

How to verify outputs

<examples>

Multi-shot learning

<anti_patterns>

Common mistakes to avoid

<security_checklist>

Security patterns

skills/{skill-name}/ ├── SKILL.md # Main entry point ├── references/ # Additional documentation │ └── api-reference.md └── templates/ # Optional templates └── example.md

Slash Command Wrapper

After creating a skill, create a lightweight slash command wrapper to make it invocable via /skill-name:

Location: ~/.claude/commands/{skill-name}.md

Template:

YAML

---
description: {Brief description of what the skill does}
argument-hint: [{argument description}]
allowed-tools: Skill({skill-name})
---

<objective>
Delegate {task} to the {skill-name} skill for: #$ARGUMENTS
</objective>

<process>
1. Use Skill tool to invoke {skill-name} skill
2. Pass user's request: #$ARGUMENTS
3. Let skill handle workflow
</process>

<success_criteria>
- Skill successfully invoked
- Arguments passed correctly to skill
</success_criteria>

The slash command's only job is routing—all expertise lives in the skill.