Summary

Project NOVA is a self-hosted AI assistant ecosystem that orchestrates over 25 specialized agents through n8n workflows and Model Context Protocol (MCP) servers. It enables users to control diverse applications, from knowledge bases and development tools to home automation, providing a centralized, extensible, and privacy-focused AI assistant capable of managing complex digital and physical environments.

How It Works

NOVA employs a sophisticated hub-and-spoke architecture. At its core, the n8n workflow engine orchestrates communication. An intelligent router agent analyzes incoming requests and conversation context to determine the most appropriate specialized agent. Over 25 domain-specific agents, each powered by the Model Context Protocol (MCP) and often running as containerized services (via provided Dockerfiles), provide distinct capabilities. This modular design allows for extensibility, enabling seamless integration with various applications and devices, from local development tools and documentation platforms to smart home systems. The use of MCP standardizes agent communication, while n8n handles the workflow logic and routing.

Quick Start & Requirements

• Primary Install/Run: Clone the repository. Deployment involves setting up n8n workflows and MCP servers, with recommended options including OpenWebUI integration for a richer user experience or a basic n8n-only setup. MCP servers are typically deployed using provided Dockerfiles and docker-compose.yml.
• Prerequisites: A running n8n instance (v1.88.0 or later is critical for native MCP support), a Docker host environment, access to a Large Language Model (via cloud APIs like OpenAI, Claude, Gemini, or a local Ollama instance), and the specific applications intended for control (e.g., Home Assistant, Gitea, Reaper, OBS Studio). For the recommended OpenWebUI setup, an instance is required, necessitating SSL/HTTPS for text-to-speech features.
• Resource Footprint: A basic setup requires approximately 4GB RAM and 2 CPU cores. For the full OpenWebUI and agent ecosystem, 16GB RAM and 8 CPU cores are recommended. Individual MCP server resource usage will vary based on the agent's function.
• Links: Repository: https://github.com/dujonwalker/project-nova

Highlighted Details

• Extensive Agent Network: Integrates over 25 specialized agents covering Knowledge Management (TriliumNext, BookStack, Paperless-NGX), Development (Forgejo, Gitea, CLI Server), Media Production (Ableton, OBS, Reaper), and Home Automation (Home Assistant).
• Conversation-Aware Intelligence: Maintains context across interactions, enabling intelligent follow-up routing and a consistent user experience across agent handoffs.
• OpenWebUI & Native n8n Integration: Leverages OpenWebUI for a modern chat interface with conversation persistence and utilizes n8n's native MCP Client Tool node, eliminating the need for community nodes.
• Diverse Use Cases: Demonstrates capabilities in complex scenarios like "Turn off living room lights and start my evening playlist" or "Check Gitea repositories for open pull requests."

Maintenance & Community

Primary community interaction and support are managed through GitHub Issues. The README does not provide direct links to community channels like Discord or Slack, nor does it detail a public roadmap.

Licensing & Compatibility

The project's README does not explicitly state a software license. This omission represents a significant barrier for potential adopters, particularly for commercial use or integration into proprietary systems.

Limitations & Caveats

The absence of a clear software license is a critical adoption blocker. The recommended hardware specifications for a full ecosystem setup are substantial (16GB RAM, 8 CPU cores). The project lists future development items as checkboxes, indicating that certain features are still under active development or planned.