Open Orchestra is a monorepo project that provides a multi-agent orchestration plugin for OpenCode. It enables users to spawn, manage, and coordinate specialized AI workers, addressing the need for more effective AI task execution by breaking down complex problems into steps handled by distinct, optimized agents (e.g., Vision, Docs, Coder). This approach leads to improved results, clearer reasoning, and parallel processing capabilities for developers within the OpenCode ecosystem.

How It Works

The project employs a hub-and-spoke architecture, centralizing coordination within the orchestrator while delegating specific tasks to specialized workers. This design leverages distinct AI models optimized for particular functions like image analysis, documentation research, or code generation. Workflows define multi-step plans executed across these workers, allowing for complex automation and clearer reasoning compared to single-AI approaches. The system is built as a plugin for the OpenCode CLI.

Quick Start & Requirements

Installation involves adding the opencode-orchestrator package via Bun: bun add opencode-orchestrator or bun add -g opencode-orchestrator. Prerequisites include the Bun runtime (v1.0.0+), the OpenCode CLI, and a configured AI provider (e.g., Anthropic API key set in ~/.config/opencode/opencode.json). Detailed setup instructions are available in the project's Quickstart Guide.

Highlighted Details

• Features six built-in worker profiles: Vision, Docs, Coder, Architect, Explorer, and Memory.
• Implements a 5-tool asynchronous task API for managing worker and workflow execution.
• Supports an optional Neo4j-backed memory system for persistent knowledge storage and context pruning.
• Utilizes dynamic port allocation and session-based isolation for workers.

Maintenance & Community

The README indicates contributions are welcome via pull requests to the main branch, referencing a CONTRIBUTING.md file. No specific details regarding maintainers, sponsorships, or community channels (like Discord or Slack) are provided in the text.

Licensing & Compatibility

The project is released under the MIT license, which is permissive and generally compatible with commercial use and linking within closed-source applications.

Limitations & Caveats

The Neo4j memory system is an optional, advanced feature. The project is specifically designed as a plugin for OpenCode, indicating tight integration and potential ecosystem dependency. No other explicit limitations or alpha/beta status indicators were found in the provided text.