Open Code Agents

A CLI tool for managing specialized AI agents that automate software development workflows. This project provides a structured way to orchestrate different AI personas (Architect, Implementer, Tester, etc.) for various development tasks.

Features

• Agent Management: Install and manage specialized AI agents
• Workflow Orchestration: Execute multi-agent workflows for common development tasks
• Interactive CLI: User-friendly command-line interface
• Scope-based Installation: User-wide or project-specific agent installations
• Configuration Management: Persistent settings and preferences

Quick Start

Installation

# Build from source
go build ./cmd/opencode-setup

# Or use the provided binary
./opencode-setup

Basic Usage

# List available agents and workflows
./opencode-setup commands list

# Execute a specific agent
./opencode-setup commands execute implementer

# Run a predefined workflow
./opencode-setup commands workflow

# Interactive agent management
./opencode-setup agents

Available Agents

Agent	Role	Description
Architect	System Design	High-level architectural planning and design decisions
Implementer	Code Development	Writes production-quality code based on specifications
Tester	Quality Assurance	Creates comprehensive test suites
Debugger	Issue Resolution	Analyzes and fixes bugs in code
Reviewer	Code Review	Evaluates code quality and adherence to standards
Refactorer	Code Improvement	Improves code structure and maintainability
Documenter	Documentation	Creates comprehensive documentation
Researcher	Investigation	Researches technologies and solutions

Predefined Workflows

New Feature Development

./opencode-setup commands workflow
# Select "new-feature"
# Provide context: feature_description="Add user authentication"

Sequence: Researcher → Architect → Implementer → Tester → Reviewer → Documenter

Bug Fix

./opencode-setup commands workflow
# Select "bug-fix"
# Provide context: bug_description="Login fails with valid credentials"

Sequence: Debugger → Implementer → Tester → Reviewer

Code Improvement

./opencode-setup commands workflow
# Select "code-improvement"
# Provide context: code_location="src/auth/"

Sequence: Refactorer → Tester → Reviewer

Installation Scopes

User Scope

• Location: ~/.config/opencode/
• Available to all projects
• Ideal for commonly used agents

Project Scope

• Location: .opencode/ in project root
• Project-specific agents and configurations
• Version controlled with project

Configuration

Configuration is stored in JSON format at:

• User: ~/.config/opencode/config.json
• Project: .opencode/config.json

Example configuration:

{
  "default_agent": "implementer",
  "workflows": {
    "my-feature": "new-feature"
  },
  "settings": {
    "log_level": "info",
    "auto_save": true,
    "show_hints": true,
    "max_history": 100
  }
}

Development

Project Structure

├── cmd/opencode-setup/     # Main CLI application
├── pkg/
│   ├── agent/              # Agent execution engine
│   ├── cli/                # CLI commands and menus
│   ├── config/             # Configuration management
│   ├── installer/          # Agent installation system
│   ├── interactive/        # Interactive UI components
│   ├── orchestrator/       # Workflow orchestration
│   └── resources/          # Embedded agent definitions
├── agents/                 # Agent definition files
└── embed/                  # Embedded resources

Building and Testing

# Build the application
go build ./cmd/opencode-setup

# Run tests
go test ./...

# Run specific package tests
go test ./pkg/agent
go test ./pkg/orchestrator

Adding New Agents

1. Create agent definition in agents/ directory
2. Follow the established template format
3. Include role, responsibilities, workflow sections
4. Update agent resource parsing if needed
5. Add tests for the new agent

Creating Custom Workflows

// Define workflow in pkg/orchestrator/workflow.go
customWorkflow := orchestrator.Workflow{
    Name:        "custom-workflow",
    Description: "Custom workflow for specific needs",
    Steps: []orchestrator.WorkflowStep{
        {
            AgentName: "researcher",
            Input:     "Research {{topic}}",
            Required:  true,
        },
        {
            AgentName: "implementer", 
            Input:     "Implement based on research: {{last_output}}",
            Required:  true,
        },
    },
}

Contributing

1. Fork the repository
2. Create a feature branch
3. Add tests for new functionality
4. Ensure all tests pass
5. Submit a pull request

License

This project is open source. See LICENSE file for details.

Support

For issues and questions:

• Check the documentation
• Review existing issues
• Create new issue with detailed description