create-atlasclaw-skill

Create AtlasClaw Skill

Guide for creating new skills that extend the AtlasClaw AI Agent's capabilities.

Quick Start Checklist

Skill Creation Progress:
- [ ] 1. Gather requirements (skill type, purpose, triggers)
- [ ] 2. Choose storage location (personal or project)
- [ ] 3. Create skill directory structure
- [ ] 4. Write SKILL.md with LLM context fields
- [ ] 5. Implement handler (for executable skills)
- [ ] 6. Test skill loading and execution

Phase 1: Gather Requirements

Before creating a skill, determine:

Phase 2: Directory Structure

Create skill at: {location}/skills/{skill-name}/

Minimal Structure (Markdown Skill)

skills/{skill-name}/
└── SKILL.md              # Required - skill metadata and documentation

Complete Structure (Executable Skill)

skills/{skill-name}/
├── SKILL.md              # Required - skill metadata
├── README.md             # Optional - extended documentation
├── scripts/              # Required for executable skills
│   ├── __init__.py
│   └── handler.py        # Main implementation
├── tests/                # Optional - test files
│   └── test_handler.py
└── references/           # Optional - reference docs
    └── api_reference.md

Phase 3: SKILL.md Template

---
name: "{skill-name}"
description: "Brief description. Trigger when user wants to {action}."
category: "{category}"
provider_type: "{provider}"           # For provider skills only
instance_required: "{true|false}"     # For provider skills only
version: "1.0.0"
author: "[email protected]"

# === LLM Context Fields (for Skill Discovery) ===
triggers:
  - action phrase 1
  - action phrase 2

use_when:
  - User intent scenario 1
  - User intent scenario 2

avoid_when:
  - Scenario when other skill is better

examples:
  - "Example user input 1"
  - "Example user input 2"

related:
  - related-skill-1
  - related-skill-2

# === Tool Registration (for executable skills) ===
tool_name: "{skill_name}"
tool_entrypoint: "scripts/handler.py:handler"
---

# {skill-name}

## Purpose

What this skill does and when to use it.

## Parameters

### Input

| Name | Type | Required | Description |
|------|------|----------|-------------|
| param1 | string | Yes | Description |
| param2 | integer | No | Description (default: 10) |

### Output

| Name | Type | Description |
|------|------|-------------|
| success | boolean | Whether operation succeeded |
| message | string | Human-readable result |
| data | object | Structured result data |

## Usage Examples

### Example 1: Basic Usage

Input:
```json
{
  "param1": "value1"
}

Output:

{
  "success": true,
  "message": "Operation completed",
  "data": { "result": "value" }
}

Error Handling

Common Errors

Related Skills

• related-skill - Description

Notes

Additional information, limitations, or considerations.