OmniNova Claw is a local-first AI agent platform and desktop control plane designed for developers and power users. It provides a unified, secure, and extensible foundation for building complex agent workflows, managing multiple LLM providers, and deploying bots across various communication channels, offering complete control via a modern desktop interface.

How It Works

The platform features a high-performance Rust core runtime (omninova-core) paired with a Tauri + React desktop interface (omninova-tauri). Its architecture is modular, with the Rust backend handling agent logic, skills, tools, LLM provider integrations, channel adapters, and an HTTP gateway. Key innovations include the "Soul System" for MBTI-driven agent personas and adaptive behavior, a three-layer memory system (Working, Episodic, Semantic/Skill) for context management, and the Agent Control Protocol (ACP) for standardized tool interaction.

Quick Start & Requirements

• Prerequisites: Rust (latest stable), Node.js (v22+). Linux requires libwebkit2gtk-4.1-dev, libappindicator3-dev, librsvg2-dev; Windows requires Visual Studio C++ Build Tools.
• Installation: Clone the repository (git clone https://github.com/omninova/claw.git), navigate to claw/omninovalclaw/apps/omninova-tauri, and run npm install.
• Running: Use npm run tauri dev for development or npm run tauri build for production builds. Artifacts are generated in apps/omninova-tauri/src-tauri/target/release/bundle/.
• Configuration: Managed via config.toml (default ~/.omninova/config.toml) or environment variables, with a setup wizard in the UI.

Highlighted Details

• Local-first AI agent platform with a Rust core and Tauri/React desktop UI.
• "Soul System" for MBTI-driven agent personas and adaptive behavior.
• Three-layer memory architecture (Working, Episodic, Semantic/Skill).
• Extensive LLM provider support (OpenAI, Anthropic, Gemini, Ollama, etc.) and multi-channel connectivity (Slack, Discord, WeChat, etc.).
• Built-in tools (file, web, PDF, Git, shell) and an extensible Skills System compatible with OpenClaw.
• Cross-platform native desktop applications (macOS, Windows, Linux).

Maintenance & Community

Automated cross-platform builds are handled via GitHub Actions. The provided README does not detail specific contributors, sponsorships, community channels (e.g., Discord/Slack), or a public roadmap.

Licensing & Compatibility

The project is licensed under the permissive MIT License, generally allowing for commercial use and integration into closed-source projects.

Limitations & Caveats

The README does not explicitly state the project's development stage (e.g., alpha/beta), though release tagging suggests potentially early versions. Specific system dependencies are required for building on Linux and Windows. No performance benchmarks are presented.