RepoChief CLI

Command-line interface for RepoChief - AI Agent Orchestration Engine

RepoChief CLI provides an easy way to orchestrate swarms of AI coding agents from your terminal.

Installation

npm install -g @liftping/repochief-cli

Or run directly with npx:

npx @liftping/repochief-cli demo --mock

Quick Start

1. Run the Demo

Try the TODO API demo that showcases a 4-agent swarm:

# Run in mock mode (no API costs)
repochief demo --mock

# Run with real AI models
repochief demo

2. Initialize a Project

Create a new RepoChief project:

repochief init my-project
cd my-project

3. Configure API Keys

Add your API keys to the .env file:

OPENAI_API_KEY=your-openai-key
ANTHROPIC_API_KEY=your-anthropic-key
GOOGLE_API_KEY=your-google-key

4. Run Tasks

Execute your AI agent tasks:

repochief run tasks/default.json

Authentication

RepoChief CLI supports cloud synchronization across multiple devices. First, authenticate with your RepoChief account:

# Login with OAuth device flow (recommended)
repochief auth login

# Login with Personal Access Token
repochief auth login --token your-pat-token

# Check authentication status
repochief auth status

# Show detailed usage info
repochief auth status --verbose

# Logout
repochief auth logout

Commands

repochief auth

Manage authentication with RepoChief cloud.

Subcommands:

• login - Authenticate using OAuth device flow or PAT
• logout - Log out from RepoChief cloud
• status - Check authentication and sync status

repochief run <task-file>

Execute AI agents with a task configuration file.

Options:

• -a, --agents <number> - Number of agents to spawn (default: 3)
• -b, --budget <amount> - Total budget in USD (default: 10)
• -m, --mock - Run in mock mode (no API calls)
• -w, --watch - Watch progress in real-time
• -o, --output <dir> - Output directory for results (default: ./output)

Example:

repochief run tasks/feature-development.json --agents 5 --budget 20 --watch

repochief init [project-name]

Initialize a new RepoChief project with templates and configuration.

Options:

• -n, --name <name> - Project name
• -t, --template <type> - Project template: basic or advanced (default: basic)

repochief demo

Run the TODO API demo to see RepoChief in action.

Options:

• -m, --mock - Run in mock mode (recommended for first try)
• -b, --budget <amount> - Budget for the demo (default: 5)

repochief agents

List available agent profiles and their capabilities.

Options:

• -j, --json - Output in JSON format
• -c, --create - Create a new agent profile (coming soon)
• -e, --edit <name> - Edit an existing agent profile (coming soon)

repochief status

Check RepoChief system status and recent sessions.

Options:

• -s, --session <id> - Check specific session status
• -c, --costs - Show cost breakdown
• -t, --tasks - Show task progress

repochief config

Manage RepoChief configuration.

Options:

• --api-keys - Configure API keys interactively

Task Configuration

Tasks are defined in JSON files with the following structure:

{
  "tasks": [
    {
      "id": "analyze-code",
      "type": "comprehension",
      "objective": "Analyze the authentication system",
      "context": ["src/auth/", "docs/"],
      "maxTokens": 50000
    },
    {
      "id": "implement-feature",
      "type": "generation",
      "objective": "Add two-factor authentication",
      "dependencies": ["analyze-code"],
      "successCriteria": [
        "SMS-based 2FA",
        "TOTP support",
        "Backward compatibility"
      ],
      "maxTokens": 80000
    }
  ]
}

Task Types

• comprehension: Understand existing code/requirements
• generation: Create new code/features
• validation: Verify correctness/quality
• exploration: Research solutions

Agent Profiles

RepoChief includes pre-configured agent profiles:

• SENIOR_DEVELOPER: GPT-4 based, complex tasks
• QA_ENGINEER: Testing and validation specialist
• SECURITY_EXPERT: Claude-3 based security analysis
• ARCHITECT: High-context system design
• JUNIOR_DEVELOPER: Cost-effective simple tasks

Environment Variables

• OPENAI_API_KEY: OpenAI API key
• ANTHROPIC_API_KEY: Anthropic (Claude) API key
• GOOGLE_API_KEY: Google AI API key
• MOCK_MODE: Set to "true" for mock mode
• DEFAULT_MODEL: Default AI model to use
• DEBUG: Enable debug logging
• REPOCHIEF_API_URL: Custom API endpoint (default: https://api.repochief.com/v1)
• REPOCHIEF_TOKEN: Personal Access Token for CI/CD environments

Multi-Device Usage

RepoChief CLI supports seamless usage across multiple devices:

Personal Devices

# On your work laptop
repochief auth login
# Device name: MacBook Pro - Work

# On your home desktop
repochief auth login
# Device name: Home Desktop - Windows

# Check all connected devices
repochief auth status --verbose

CI/CD Integration

For automated environments, use Personal Access Tokens:

# Generate PAT from web dashboard
# Then in CI/CD:
export REPOCHIEF_TOKEN="rcp_xxxxxxxxxxxxxxxx"
repochief run tasks/ci-validation.json

Device Management

# Logout from current device only
repochief auth logout

# Logout from all devices
repochief auth logout --all-devices

# View device-specific usage
repochief auth status --verbose

Examples

Basic Code Review

repochief run examples/code-review.json --agents 2 --budget 5

Feature Development

repochief run examples/feature-dev.json --agents 4 --budget 20 --watch

Security Audit

repochief run examples/security-audit.json --agents 3 --budget 15

Troubleshooting

"No API keys found"

Make sure to set up your API keys in the .env file or as environment variables.

"Budget exceeded"

Increase the budget with the -b flag or use mock mode for testing.

"Cannot find repochief-core"

Install the core package:

npm install @liftping/repochief-core

Security

• Never commit API keys to the repository
• Use environment variables for sensitive configuration
• Report security vulnerabilities to: [email protected]

Contributing

See the main RepoChief repository for contribution guidelines.

License

MIT