---
name: ai-eng/init
description: Initialize ai-eng-system configuration and project setup
agent: plan
version: 1.0.0
inputs:
  - name: interactive
    type: boolean
    required: false
    description: Interactive configuration setup
  - name: overwrite
    type: boolean
    required: false
    description: Overwrite existing config file
outputs:
  - name: config_file
    type: file
    format: YAML
    description: .ai-eng/config.yaml with default or custom settings
---

Default output: return only the result, blockers, and required evidence. Omit preambles, process narration, repeated context, confidence scores, and follow-up offers. Use at most five bullets unless a required artifact or schema needs more.

# Initialize Command

Initialize ai-eng-system for: $ARGUMENTS

> **Project Setup**: Initialize configuration files and development environment

Set up ai-eng-system configuration with defaults or interactive prompts. Ensures proper project structure and development environment readiness.

## Quick Start

```bash
# Initialize with defaults
/ai-eng/init

# Interactive setup
/ai-eng/init --interactive

# Force overwrite existing config
/ai-eng/init --overwrite

# Combined options
/ai-eng/init --interactive --overwrite
```

## Options

| Option | Description |
|--------|-------------|
| `-i, --interactive` | Interactive configuration setup with prompts |
| `--overwrite` | Overwrite existing configuration file |
| `-v, --verbose` | Enable verbose output |

## Configuration Options

### Interactive Mode Prompts

When `--interactive` is specified, you'll be prompted for:

1. **Default Model Selection**
   - Choose preferred AI model (claude-3.5-sonnet, gpt-4, etc.)
   - Set model parameters (temperature, max tokens)

2. **Project Type**
   - Web application (React/Vue/Angular)
   - Backend service (Node.js/Python/Go)
   - Full-stack application
   - CLI tool
   - Library/Package

3. **Development Preferences**
   - Testing framework preference
   - Code style and formatting
   - Build system choices
   - CI/CD platform

4. **Quality Gates**
   - Enable code review automation
   - Set performance thresholds
   - Security scanning preferences

### Default Configuration

Without interactive mode, creates `.ai-eng/config.yaml` with sensible defaults:

```yaml
# AI Engineering System Configuration
version: "1.0"
project:
  name: "ai-eng-project"
  type: "full-stack"
  root: "."
  
models:
  default: "claude-3.5-sonnet"
  temperature: 0.7
  max_tokens: 4096
  
development:
  test_framework: "bun"
  linter: "biome"
  formatter: "biome"
  
quality:
  code_review: true
  security_scan: true
  performance_threshold: 500
  
paths:
  source: "src/"
  tests: "tests/"
  docs: "docs/"
  build: "dist/"
  
logging:
  level: "INFO"
  file: ".ai-eng/logs/ai-eng.log"
  max_size: "10MB"
```

## Initialization Steps

### Phase 1: Environment Setup

1. **Create Directory Structure**
   ```bash
   .ai-eng/
   ├── config.yaml
   ├── logs/
   └── checkpoints/
   ```

2. **Detect Project Context**
   - Scan for existing configuration files
   - Identify project type and framework
   - Determine development environment

3. **Configuration Generation**
   - Apply defaults or user preferences
   - Validate configuration schema
   - Create or update config file

### Phase 2: Asset Verification

1. **Verify CLI Installation**
   - Check if CLI is globally available
   - Validate PATH configuration
   - Test basic functionality

2. **Plugin Registration**
   - Register with Claude Code if detected
   - Register with OpenCode if detected
   - Validate plugin permissions

3. **Development Environment**
   - Check for required tools (bun, node, etc.)
   - Verify build system compatibility
   - Validate test framework setup

### Phase 3: Integration Setup

1. **Git Hooks Configuration**
   - Install pre-commit hooks
   - Configure commit message validation
   - Set up automated testing

2. **IDE Integration**
   - Create VS Code workspace settings
   - Configure extensions recommendations
   - Set up debug configurations

3. **Documentation Generation**
   - Create initial project documentation
   - Generate API documentation stubs
   - Set up contribution guidelines

## Quality Validation

After initialization, verify:

- [ ] `.ai-eng/config.yaml` exists and is valid YAML
- [ ] CLI commands respond correctly
- [ ] Build system recognizes the configuration
- [ ] Test framework integration works
- [ ] Git hooks are properly installed
- [ ] Documentation files are created

## Troubleshooting

### Common Issues

1. **Permission Errors**
   ```bash
   # Fix: Ensure write permissions
   chmod 755 .ai-eng/
   ```

2. **Missing Tools**
   ```bash
   # Fix: Install required dependencies
   bun install
   npm install -g @ai-eng-system/cli
   ```

3. **Configuration Conflicts**
   ```bash
   # Fix: Remove existing config and reinitialize
   rm .ai-eng/config.yaml
   /ai-eng/init --overwrite
   ```

4. **PATH Issues**
   ```bash
   # Fix: Verify global installation
   which ai-eng
   export PATH="$PATH:$(npm bin -g)"
   ```

Focus on:
- **Reliability**: Ensure initialization works in diverse environments
- **Validation**: Verify all components are properly configured
- **Documentation**: Provide clear guidance for next steps
- **Error Handling**: Gracefully handle configuration conflicts

$ARGUMENTS