AI DevKit CLI

A powerful command-line tool for creating AI-powered development projects with pre-configured team role templates.

Features

• 🚀 Quick project setup with team-specific configurations
• 👥 Multiple role templates (Developer, QC, PM, BA, Product Owner)
• 📝 Customizable agent templates for different team roles
• 🎯 Flexible role selection for different project needs
• 📦 Ready to use with npx - no global installation required
• 🔄 Automated releases with semantic versioning

Installation

No installation required! Use directly with npx:

npx @trinitytechlab/ai-devkit new project --team=your-team-name

Or install globally:

npm install -g @trinitytechlab/ai-devkit
ai-devkit new project --team=your-team-name

Usage

Basic Command Structure

npx ai-devkit new project --team=<team-name> [--roles=<role1,role2,...>]

Examples

Create a project for "linkify" team with all roles:

npx @trinitytechlab/ai-devkit new project --team=linkify

Create a project with specific roles only:

npx @trinitytechlab/ai-devkit new project --team=linkify --roles=dev,qc

Create a project with custom role combination:

npx @trinitytechlab/ai-devkit new project --team=linkify --roles=dev,pm,ba

Available Roles

| Role | Description | Template File | |------|-------------|---------------| | dev | Developer | dev.md | | qc | Quality Control | qc.md | | pm | Project Manager | pm.md | | ba | Business Analyst | ba.md | | po | Product Owner | po.md |

Project Structure

When you create a new project, the following structure is generated:

ai-your-team-project/
├── dev/
│   └── AGENTS.md   # Developer agent template
├── qc/
│   └── AGENTS.md   # QC agent template
├── pm/
│   └── AGENTS.md   # PM agent template
├── ba/
│   └── AGENTS.md   # BA agent template
├── po/
│   └── AGENTS.md   # PO agent template
└── README.md       # Project documentation

Agent Templates

Each role template includes:

• Role Description: Clear definition of responsibilities
• Core Responsibilities: Key tasks and duties
• Guidelines: Best practices and standards
• Tools & Technologies: Relevant tools for the role
• Collaboration: How to work with other team members
• Quality Standards: Criteria for success

Customization

After creating a project:

1. Navigate to your project directory ai-{team}-project/
2. Edit the agent templates in each role's folder (e.g., dev/AGENTS.md)
3. Customize the content according to your team's specific needs
4. Use the templates as prompts for AI assistants or guidelines for team members

Commands

new project

Creates a new AI development project with specified team and roles.

Options:

• --team <name>: Team name (required)
• --roles <roles>: Comma-separated list of roles (default: all roles)

Interactive Mode: If you don't provide the --team option, the CLI will prompt you for the team name.

Development

Prerequisites

• Node.js 14.0.0 or higher
• npm or yarn

Setup

git clone <repository-url>
cd ai-devkit-cli
npm install
npm link  # For local development

Testing

# Test the CLI locally
node bin/cli.js new project --team=test-team --roles=dev,qc

# Test with npx (after publishing)
npx @trinitytechlab/ai-devkit new project --team=test-team --roles=dev,qc

Contributing

We use Conventional Commits for commit messages and Semantic Release for automated versioning.

Commit Convention

Please use the following format for commit messages:

<type>[optional scope]: <description>

[optional body]

[optional footer(s)]

Types:

• feat: A new feature (triggers minor version)
• fix: A bug fix (triggers patch version)
• docs: Documentation only changes (triggers patch version)
• refactor: Code refactoring (triggers patch version)
• perf: Performance improvements (triggers patch version)
• test: Adding or updating tests (no release)
• chore: Maintenance tasks (no release)
• ci: CI/CD changes (no release)

Examples:

feat: add support for custom template directory
fix: resolve template path resolution on Windows
docs: update installation instructions
refactor: improve error handling in new-project command

Development Process

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Make your changes following the coding standards
4. Write tests for new functionality
5. Commit your changes using conventional commits
6. Push to the branch (git push origin feature/amazing-feature)
7. Open a Pull Request

Release Process

Releases are automated using semantic-release:

• Patch: Bug fixes, documentation, refactoring
• Minor: New features
• Major: Breaking changes (add BREAKING CHANGE: in commit footer)

The CI/CD pipeline will automatically:

1. Run tests on Node.js 16, 18, and 20
2. Generate changelog
3. Create GitHub release
4. Publish to npm registry

NPM Token Setup (For Maintainers)

To enable automated publishing, you need to setup an NPM_TOKEN secret:

1. Create NPM Automation Token:

   # Login to npm
   npm login
      
   # Create automation token (bypasses 2FA)
   npm token create --type=automation

2. Add to GitHub Secrets:

   • Go to repository Settings → Secrets and variables → Actions
   • Add new repository secret: NPM_TOKEN
   • Paste the automation token value

3. Token Permissions:

   • Automation tokens bypass 2FA requirements
   • They have the same permissions as your npm account
   • Store them securely and rotate regularly

License

This project is licensed under the MIT License - see the LICENSE file for details.

Support

For support and questions:

• Create an issue on GitHub
• Contact the TrinityTech team

Made with ❤️ by TrinityTech