.cursorrules git:20260701.14a4958A

Immutable. This exact content is served forever at /api/v1/blob/e0b76a5896555455.

# Wondel.ai Skills Repository

You are working in a collection of agent skills for Claude, Claude Code, Claude Cowork, Codex, Cursor, OpenClaw, Hermes Agent and other agentskills.io-compatible agents.

## Repository Purpose

This repository contains specialized domain knowledge skills for AI agents. Each skill provides frameworks and methodologies for specific use cases (UX design, marketing, product strategy, etc.).

## Strict Requirements

### When Modifying Skills

1. **ALWAYS bump version numbers** in SKILL.md frontmatter when making changes:
   - MINOR bump (e.g., 1.0.0 → 1.1.0) for new features or significant enhancements
   - PATCH bump (e.g., 1.0.1 → 1.0.2) for bug fixes, typos, or small clarifications

2. **YAML frontmatter is REQUIRED** - Every SKILL.md must have:
   ```yaml
   ---
   name: skill-name
   description: What the skill does and when to use it (max 1024 chars, include trigger phrases)
   license: MIT
   metadata:
     author: wondelai
     version: "x.y.z"
   ---
   ```

3. **Name constraints**:
   - Max 64 characters
   - Lowercase letters, numbers, hyphens only
   - No start/end hyphens, no consecutive hyphens
   - MUST match parent directory name exactly

4. **Description requirements**:
   - Max 1024 characters
   - Include specific trigger phrases and keywords
   - Describe BOTH what it does AND when to use it

### Directory Structure

```
skills/
├── .claude-plugin/
│   └── marketplace.json   # Plugin marketplace configuration
├── {skill-name}/
│   ├── SKILL.md           # Main skill file (YAML frontmatter + markdown)
│   └── references/        # Optional: supporting reference files
│       └── *.md
└── README.md              # Skill catalog
```

### Adding New Skills

1. Create folder: `{skill-name}/`
2. Create `SKILL.md` with valid YAML frontmatter
3. Add to `README.md` with description and examples
4. Add to `.claude-plugin/marketplace.json` in appropriate collection

### Code Style

- Use clear, semantic markdown
- Keep main SKILL.md concise (< 500 lines recommended)
- Move detailed content to `references/*.md` files
- Use progressive disclosure: metadata → instructions → resources

### Quality Standards

- Instructions should be actionable and specific
- Include examples where helpful
- Focus on practical application, not theory
- Write for AI agents as the primary audience

## Commit Policy

Commit directly to main branch - no PRs required.

## OpenAgentSkills Compliance

All skills MUST comply with the OpenAgentSkills specification (https://agentskills.io/specification). This ensures cross-platform compatibility with Claude Code, skills.sh, and other agentskills.io-compatible agents.