Skip to content
Back to skills

Skill Creation

ASecurity

Teach Claude a new permanent skill based on user-provided resources, articles, or documentation. Trigger this skill when the user asks you to "learn a new skill" or provides a resource and says "learn this".

  • 6 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agentsgoreactdocumentation

Security analysis

A100/100

Scanned November 11, 2026

npx -y skills add DaraBoth/skilljmskill --skill skill-creation --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Skill Creation?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Skill Creation
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/daraboth-skill-creation/badge)](https://www.skillsdirectory.com/skills/daraboth-skill-creation)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: skill-creation
description: Teach Claude a new permanent skill based on user-provided resources, articles, or documentation. Trigger this skill when the user asks you to "learn a new skill" or provides a resource and says "learn this".
---

# Skill: Skill Creation

> **Purpose**: Teach Claude a new permanent skill based on user-provided resources, articles, or documentation.

## When to Use
- You find a great tutorial or guide and want Claude to remember its patterns
- You want to standardize a new team workflow or coding practice
- You want to expand Claude's knowledge base in the `skills/` directory

---

## How to Teach Claude a New Skill

### Step 1: Provide the Resource
Give Claude the raw material. This can be:
- Pasting the text of an article
- Providing a link or asking Claude to search the web
- Providing a local file path to documentation

### Step 2: Prompt Claude to Learn
Use this exact prompt format to trigger the skill creation:
```
Learn a new skill from this resource: [link/text/file]
Name the skill: [Skill Name]
Category: [coding | personal | language-learning]
```

---

## Claude's Processing Workflow

When the user asks you to learn a new skill, you MUST follow this exact sequence:

### 1. Analyze and Extract
- Read the provided resource thoroughly.
- Extract the core principles, workflows, and best practices.
- Strip out fluff, marketing speak, and long historical context.

### 2. Format
Structure the extracted knowledge using the **Standard Skill Format** (see below). Ensure it strictly adheres to the ADHD-friendly design principles (scannable, short paragraphs, example-driven).

### 3. Save
Create a new `.md` file in the appropriate `skills/[category]/` directory. (e.g., `skills/coding/react-hooks.md`).

### 4. Register
Update `CLAUDE.md` to add the new skill name alphabetically to the "Available skills" list.

### 5. Confirm
Reply to the user with a short summary of what was learned and confirm that the skill is now active.

---

## Standard Skill Format

Every new skill MUST follow this structure to remain consistent:

```markdown
# Skill: [Name]

> **Purpose**: [One sentence explaining why this skill exists]

## When to Use
- [Scenario 1]
- [Scenario 2]

---

## Core Concepts
### [Concept 1]
- Bullet points only
- Keep it punchy

## Examples
### ✅ Good Example
\`\`\`language
// Clean, readable code or clear text
\`\`\`

### ❌ Bad Example
\`\`\`language
// What to avoid
\`\`\`

## Best Practices
1. **Rule 1** — Explanation
2. **Rule 2** — Explanation

## Common Mistakes
| Mistake | Fix |
|---------|-----|
| [Mistake] | [Fix] |
```

---

## Rules for Skill Creation

### 1. ADHD-Friendly Formatting
- **No walls of text**: Use short paragraphs (max 2-3 sentences).
- **Highly scannable**: Use headers, bullet points, and bold text for keywords.
- **Action-oriented**: Focus on *how to do it*, not just *what it is*.

### 2. Practicality Over Theory
- Include code snippets or concrete examples for everything.
- If the source material lacks examples, generate practical ones yourself.

### 3. Modularity
- One concept per file. If a skill gets too large, split it into sub-skills.
- Stay objective. Focus on facts and tested workflows, not unverified opinions.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.