nuwax

Nuwax AI provides a private Agentic AI platform that you can deploy locally using the official nuwax-cli tool. The system is designed to manage a main local service and optional Agent Computer (Sandbox) deployments, allowing you to orchestrate private AI agents across one or more servers. Once running, you can start, stop, and monitor the docker-based services, perform backups, and upgrade deployments through a set of convenient CLI commands. The solution emphasizes Docker-based deployment with a one-click installation script to simplify setup, and it includes management commands for service lifecycle, backups, and upgrades. Use these capabilities to deploy and manage private AI agents and their sandboxes securely on your infrastructure.

Prerequisites:

• A machine with Docker and Docker Compose installed (Docker Desktop on macOS/Windows or Docker Engine on Linux).
• Basic familiarity with running shell commands.

Installation steps:

1. Prepare your environment

• Ensure your OS meets the platform requirements (Linux, macOS, or Windows) and Docker is running.

2. Install Docker (if not already installed)

• Follow the official Docker installation guide for your platform. For quick verification, run:

docker --version
docker compose version

3. Install Nuwax via the one-click Docker script (recommended)

• Run the one-click installer to set up Docker and related components configured for Nuwax:

bash <(wget -qO- https://xuanyuan.cloud/docker.sh)

4. Deploy the Main Service using nuwax-cli

• After Docker is installed, use the nuwax-cli to deploy the main service and any Agent Computer (Sandbox) deployments as needed. Typical commands include:

# Start the main service (via nuwax-cli)
./nuwax-cli docker-service start

# Check status
./nuwax-cli docker-service status

# Manage backups
./nuwax-cli auto-backup run
./nuwax-cli list-backups
./nuwax-cli rollback [BACKUP_ID]

# Upgrade deployment
./nuwax-cli auto-upgrade-deploy run

5. Optional: Verify Docker environment

• Confirm Docker is properly installed and accessible by running a test container:

docker run hello-world

Notes:

• The installation script supports multiple Linux distributions and mirrors, with automatic configuration to optimize network performance in your region.
• The deployment relies on Docker and Docker Compose; ensure you have the appropriate permissions to run Docker commands (you may need sudo on some systems).

Tips and common issues:

• Docker permissions: If you encounter permission errors when running docker ps, consider adding your user to the docker group or run commands with sudo.
• Platform support: The Quick Start emphasizes Linux (Ubuntu 22.04+), macOS, and Windows support is in progress. For best results on macOS, use OrbStack or Docker Desktop.
• Agent Computer (Sandbox): You can deploy sandboxed agent environments on multiple servers to enable distributed private AI agents; each server must have Docker and Docker Compose installed.
• Mirror acceleration: For mainland China users, enable domestic mirrors during installation to speed up downloads.
• Environment variables: The deployment typically uses Docker images; if you have custom configurations, you may extend the docker run command with environment variables (e.g., -e VAR=value) as needed.