NeuralAgent is an AI-powered desktop assistant designed for task automation, capable of interacting with your computer like a human user. It targets power users and developers seeking to automate repetitive digital tasks, offering a significant productivity boost beyond conversational AI.

How It Works

NeuralAgent employs a modular architecture with a FastAPI backend, an Electron desktop application featuring a React frontend, and a Python-based AI agent daemon. It leverages pyautogui for direct desktop interaction (typing, clicking) and supports various LLM providers (OpenAI, Anthropic, Gemini, Ollama, etc.) for task planning and execution. The system is designed for extensibility, allowing for custom agents and model integrations.

Quick Start & Requirements

• Installation: Requires Python (>=3.9), PostgreSQL (>=13), and Node.js/npm (>=18/9).
• Setup: Involves setting up a PostgreSQL database, configuring environment variables (.env files for backend and frontend), installing Python and Node.js dependencies, and running backend and frontend servers separately.
• Resources: Requires two terminal windows for backend and desktop processes.
• Docs: Website, [Discord](Join NeuralAgent Discord)

Highlighted Details

• Supports multiple LLM providers including OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Gemini, and Ollama.
• Features modular agents for planning, classification, suggestion, titling, and summarization.
• Includes multimodal capabilities (text + vision).
• Offers background automation via WSL, currently Windows-only.

Maintenance & Community

The project welcomes contributions via pull requests. Community support is available through Discord.

Licensing & Compatibility

Licensed under the MIT License. This license permits commercial use and integration with closed-source projects, but users are advised to test responsibly due to the nature of the automation.

Limitations & Caveats

Background automation is currently limited to Windows via WSL. The project explicitly states "Use at your own risk" due to its direct system interaction capabilities.