Operator OS

The ultra-lightweight AI agent operating system. Run autonomous agents on hardware starting at $10 with a memory footprint under 10MB.

Operator OS is a high-performance personal AI agent framework written in Go. Deploy a single self-contained binary to any device — from a Raspberry Pi to a cloud server — and connect your agents to the LLMs and messaging channels you already use.

Quick Start

Get your agent online in under five minutes

Installation

Download binaries or build from source

Configuration

Configure models, channels, and tools

CLI Reference

Every command, flag, and option

Why Operator OS

Ultra-lightweight

Under 10MB RAM. Cold starts in under 1 second, even on a single-core 0.6GHz processor.

Single binary

Deploys as one self-contained executable across RISC-V, ARM64, and x86_64.

Persistent memory

Structural long-term memory that carries context across sessions and reboots.

Multi-channel

Slack, Discord, Telegram, WhatsApp, DingTalk, Feishu, LINE, WeCom, and QQ — out of the box.

Any LLM

OpenAI, Anthropic, Google Gemini, Groq, Ollama, DeepSeek, and more — zero code changes.

Extensible

Expand capabilities with MCP servers and installable skills from the ClawHub registry.

Get started in three steps

1. Install Operator OS

   Download the binary for your platform or build from source:

   # Build from source (requires Go 1.21+)
   git clone https://github.com/operatoronline/standard-operator.git
   cd standard-operator
   make deps && make build
   
2. Initialize your workspace

   Run the interactive onboarding wizard to create your configuration:

   operator onboard
   

   This creates ~/.operator/config.json with your chosen LLM and channels.

3. Start the Gateway

   Bring your agent online and connect it to your messaging channels:

   operator gateway
   

   You can now message your agent through Slack, Telegram, Discord, or any configured channel.

Explore the docs

AI Model providers

Configure OpenAI, Anthropic, Gemini, Ollama, and more

Messaging channels

Connect Slack, Telegram, Discord, WhatsApp, and others

Built-in tools

Shell execution, web search, filesystem, cron scheduling, and hardware I/O

MCP integration

Connect external Model Context Protocol servers

Skills marketplace

Discover and install skills from the ClawHub registry

Docker deployment

Run Operator OS fully containerized