CACI (Code Assistant Configuration Interface) - Intelligent CLI tool for configuring Claude Code projects
npm install code-assistant-config-interfaceCACI (Code Assistant Configuration Interface) - An intelligent CLI tool for configuring Claude Code projects.
CACI (Code Assistant Configuration Interface) is an npm package that provides an interactive CLI interface to automate the configuration of Claude Code projects. It intelligently analyzes project requirements and selects the most relevant agents, commands, MCPs, and hooks from a large pool of available components.
- Interactive CLI Interface: Easy-to-use command-line interface similar to the BMAD-method installer
- AI-Powered Component Selection: Uses OpenRouter API with automatic key creation to intelligently recommend components based on project requirements
- Component Analysis: Parses a large pool of components (102+ agents) and recommends the most relevant ones
- Colorful Output: User-friendly interface with colored output for better readability
- Comprehensive Testing: Full test coverage for all functionality
No installation required! Run directly with npx:
``bash`
npx code-assistant-config-interface configure
Or install globally:
`bash`
npm install -g code-assistant-config-interface
`bashRun configuration (recommended)
npx code-assistant-config-interface configure
Commands
-
configure: Start the configuration process for a new project
- update: Update the existing configuration
- reset: Restore the previous configuration
- --help: Display help informationRequirements
To use the AI-powered component recommendation feature:
1. The tool will automatically prompt you to authenticate with OpenRouter when needed
2. It uses OpenRouter's auto-create key format for seamless authentication
3. No manual API key setup required - authentication is handled through OAuth flow
Development
$3
- Node.js (v14 or higher)
- npm
$3
`bash
git clone
cd caci
npm install
`$3
`bash
npm run build
`$3
`bash
npm test
`$3
`bash
npm start
`Project Structure
`
src/
├── cli/ # CLI interface and commands
├── analyzer/ # Component analysis and AI recommendations
│ ├── index.ts # Data models and interfaces
│ ├── parser.ts # Component JSON parsing
│ ├── questions.ts # Interactive question flow
│ ├── requirementCollector.ts # CLI prompts for requirements
│ ├── ai-recommender.ts # AI-powered component recommendation
│ └── display.ts # Recommendation display
└── types/ # TypeScript typestests/
├── cli/ # CLI interface tests
└── analyzer/ # Component analysis tests
``1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a pull request
MIT
- Uses OpenRouter API for AI-powered recommendations
- Built with Commander.js for CLI interface
- Uses Chalk for colorful terminal output
- Integrates with Claude Code project configuration system