1. GATHER: Collect project context
   - What type of project? (web app, CLI, API, mobile, etc.)
   - What technologies? (React, Python, Node, etc.)
   - What workflows? (git, CI/CD, testing, etc.)
   - What integrations? (databases, APIs, services)
   - Team size and experience level?

2. IDENTIFY: Determine needs by category
   - Development personas needed (agents)
   - Repetitive actions to automate (commands)
   - Environment configurations (settings)
   - Event-driven automations (hooks)
   - External service connections (MCPs)
   - Complex capability packages (skills)

3. MATCH: Map needs to existing components
   - Reference the component library (see references/component-library.md)
   - Identify gaps requiring custom components
   - Note synergies between components

## Recommended Tech Stack for [Project Name]

### 🤖 Agents (n selected)
| Component | Why | Synergies |
|-----------|-----|-----------|
| [name] | [reasoning] | Works with [x, y] |

### ⚡ Commands (n selected)
...

### ⚙️ Settings (n selected)
...

### 🪝 Hooks (n selected)
...

### 🔌 MCPs (n selected)
...

### 🎨 Skills (n selected)
...

### Effectiveness Score: X/10
- Coverage: How well does this stack cover project needs?
- Synergy: How well do components work together?
- Efficiency: Are there redundancies or gaps?

### Installation Command
\`\`\`bash
npx organized-ai install [component-list]
\`\`\`

.claude/agents/
└── agent-name.md

---
name: agent-name
description: When to invoke this agent and what it specializes in
---

# Agent Name

## Role
[Define the persona and expertise]

## Responsibilities
- [Key responsibility 1]
- [Key responsibility 2]

## Guidelines
- [How the agent should behave]
- [What it should prioritize]

## Tools & Patterns
- [Preferred tools and workflows]

.claude/commands/
└── command-name.md

---
name: command-name
description: What this command does when invoked
---

# Command: /command-name

## Purpose
[What problem this solves]

## Usage
\`\`\`
/command-name [arguments]
\`\`\`

## Behavior
1. [Step 1]
2. [Step 2]
3. [Step 3]

## Examples
- `/command-name file.ts` - [result]

.claude/settings.json

{
  "permissions": {
    "allow": ["npm test", "npm run build"],
    "deny": ["rm -rf"]
  },
  "preferences": {
    "autoCommit": false,
    "lintOnSave": true
  },
  "environment": {
    "terminal": "zsh",
    "editor": "vscode"
  }
}

.claude/hooks/
└── hook-name.md

---
name: hook-name
description: What this hook does and when it triggers
trigger: [PreToolUse|PostToolUse|PreMessage|PostMessage]
tools: [Edit|Write|Bash|*]  # Which tools trigger this
---

# Hook: hook-name

## Trigger Conditions
- Event: [when this fires]
- Filter: [which operations]

## Behavior
[What the hook does]

## Implementation
\`\`\`bash
#!/bin/bash
# Hook script
[commands]
\`\`\`

mcp-name/
├── package.json
├── src/
│   └── index.ts
└── README.md

{
  "mcpServers": {
    "server-name": {
      "command": "npx",
      "args": ["-y", "package-name"],
      "env": {
        "API_KEY": "${API_KEY}"
      }
    }
  }
}

skill-name/
├── SKILL.md (required)
├── scripts/     (optional - executable code)
├── references/  (optional - documentation)
└── assets/      (optional - templates, images)

---
name: skill-name
description: Comprehensive description of what this skill does and when to use it. Include trigger phrases and contexts.
---

# Skill Name

## Overview
[What this skill enables]

## Workflow
1. [Step 1]
2. [Step 2]

## Resources
- scripts/tool.py - [purpose]
- references/guide.md - [when to load]

Project Need          | Component          | Coverage
---------------------|--------------------|---------
[need 1]             | [component]        | ✅/⚠️/❌
[need 2]             | [component]        | ✅/⚠️/❌

Analyze this project and recommend a tech stack:
- Project type: [type]
- Technologies: [list]
- Key workflows: [list]

Create a [agent|command|setting|hook|mcp|skill] for:
- Purpose: [what it does]
- Trigger: [when it activates]
- Behavior: [how it works]

Generate installation command for:
- [component 1]
- [component 2]
- ...

# Single component
npx organized-ai install [type]:[category]/[name]

# Multiple components
npx organized-ai install agent:dev-team/frontend command:git/commit hook:auto/lint-on-save

# From stack file
npx organized-ai install --stack my-stack.json

python3 /mnt/skills/examples/skill-creator/scripts/package_skill.py <skill-folder> /mnt/user-data/outputs