What Is NanoBot?
NanoBot is an ultra-lightweight personal AI assistant created by the HKU Data Science Lab (HKUDS). Inspired by OpenClaw, it delivers core agent functionality in just ~4,000 lines of code — that is 99% smaller than Clawdbot's 430,000+ lines. Despite its tiny footprint, NanoBot is remarkably capable.
With 19,300+ GitHub stars and 2,800+ forks, NanoBot has quickly become one of the most popular open-source AI agent projects. Its clean, readable codebase makes it ideal for businesses that want to understand and customize their AI tools.
NanoBot vs. OpenClaw: A Comparison
| Feature | OpenClaw | NanoBot |
|---|---|---|
| Codebase Size | 430,000+ lines | ~4,000 lines |
| Language | Multiple | Python |
| Setup Complexity | Medium-High | Low |
| Resource Usage | High | Low |
| Chat Platforms | 5+ | 9+ (Telegram, Discord, WhatsApp, Slack, Email, QQ, Feishu, Mochat, DingTalk) |
| MCP Support | Yes | Yes |
| Local LLM Support | Limited | Yes (vLLM) |
| Research-Friendly | No | Yes |
Key Features
Ultra-Lightweight: Just ~4,000 lines means faster startup, lower resource usage, and quicker iterations. You can read and understand the entire codebase in an afternoon. Multi-Platform Chat: NanoBot supports 9 chat platforms out of the box — more than any other AI agent. Whether your team uses Slack, Discord, Telegram, or even QQ, NanoBot integrates seamlessly. Research-Ready: The clean architecture makes it easy to modify, extend, and experiment. Perfect for businesses that want to build custom AI workflows. Lightning Fast: Minimal footprint means sub-second startup times and efficient resource usage — ideal for running on modest hardware.Quick Setup (5 Minutes)
Step 1: Install NanoBot
Choose your preferred installation method:
# Option A: Install with uv (recommended)
uv tool install nanobot-ai
# Option B: Install from PyPI
pip install nanobot-ai
# Option C: Install from source
git clone https://github.com/HKUDS/nanobot.git
cd nanobot
pip install -e .
Step 2: Initialize
nanobot onboard
Step 3: Configure Your LLM
Edit ~/.nanobot/config.json:
{
"providers": {
"openrouter": {
"apiKey": "sk-or-v1-your-key"
}
},
"agents": {
"defaults": {
"model": "anthropic/claude-sonnet-4"
}
}
}
Step 4: Start Chatting
nanobot agent
That is it — you have a working AI assistant in under 5 minutes.
Business Use Cases
24/7 Market Analysis: NanoBot can monitor stock prices, crypto markets, competitor pricing, and industry news around the clock. Set up alerts for significant changes and receive summaries via your preferred chat app. Full-Stack Development: Need a quick prototype or a bug fix? NanoBot can write code, run tests, and deploy applications. It supports multiple programming languages and frameworks. Smart Daily Routine Management: Automate your daily standup summaries, schedule meetings, set reminders, and manage task lists — all through natural language commands. Personal Knowledge Assistant: Feed NanoBot your company documentation, SOPs, and knowledge base. It becomes a searchable, conversational interface to your business knowledge.Connecting to Telegram
Setting up Telegram is the easiest way to start:
- Search for @BotFather on Telegram
- Send /newbot and follow the prompts
- Copy the bot token
- Add to your config and run
nanobot gateway
You can now message your AI assistant from anywhere — your phone, tablet, or desktop.
How Simple AI Bots Enhances NanoBot
While NanoBot is designed to be simple, businesses often need custom integrations, security hardening, and ongoing support. Simple AI Bots offers managed NanoBot deployments that include:
- Custom workflow automation tailored to your business
- Integration with your existing tools (CRM, ERP, accounting)
- Security hardening and compliance monitoring
- 24/7 monitoring and support
