What is Nanobot used for?

Nanobot is a small, self-hosted Python framework for personal AI agents with a WebUI, long-term memory, MCP tools and connections to chat apps such as Telegram, Discord and Slack. It is a good fit for a personal assistant reachable from chat apps, scheduled automations run by an AI agent and prototyping agent workflows locally.

Is Nanobot open source?

Yes. Nanobot is open source under the MIT licence. Its source code is on GitHub at HKUDS/nanobot and is written mainly in Python.

Yes. Nanobot is open source, so the software itself is free to use.

Can I self-host Nanobot?

Yes. Nanobot can be self-hosted on your own server or infrastructure; there is no official hosted version.

What are some alternatives to Nanobot?

Similar open-source tools in the AI Tools category include QwenPaw, GPT Researcher and Poco Claw. SaaS products in the same category include Aera Browser, August and CalmPilot.

Is Nanobot actively maintained?

Yes. The most recent commit to Nanobot was on 2 October 2026, and the latest release is v0.3.5, published on 15 September 2026. The project has 49k stars on GitHub.

Nanobot: open-source AI tools

About Nanobot

nanobot is an open-source, self-hosted framework for running a personal AI agent, written in Python. The agent can work from a browser WebUI, a terminal or chat apps, and the project emphasizes a small, readable core rather than a heavy platform.

Under one roof it offers tools, persistent memory, MCP integrations, model routing with fallbacks, delegation between agents, scheduled automation and an OpenAI-compatible API. Built-in tools cover files, shell, web search and fetch, cron, image generation and subagents, and channels include Telegram, Discord, Slack, WeChat, Email, Mattermost, Linear, Feishu and Teams. Sessions keep history, and long-running goals persist through a memory feature called Dream. Langfuse can be configured for tracing, and a Python SDK is available for integrations.

nanobot is MIT licensed and aimed at people who want an agent that lives on their own machine or server and talks to them through the apps they already use. The documentation includes a start-without-technical-background guide, deployment instructions for running it as a service or in the cloud, and architecture notes for contributors.

Key features

Runs in a WebUI, terminal or chat apps
Long-term memory across sessions
MCP tool integration
Multi-agent delegation and scheduled automation
OpenAI-compatible API and Python SDK
Connectors for Telegram, Discord, Slack and email

Good fit for

→A personal assistant reachable from chat apps
→Scheduled automations run by an AI agent
→Prototyping agent workflows locally

Built with: Python
Tags: ai-agent; llm; mcp; chatops; self-hosted; python; automation; multi-agent

Nanobot: questions and answers

What is Nanobot used for?: Nanobot is a small, self-hosted Python framework for personal AI agents with a WebUI, long-term memory, MCP tools and connections to chat apps such as Telegram, Discord and Slack. It is a good fit for a personal assistant reachable from chat apps, scheduled automations run by an AI agent and prototyping agent workflows locally.
Is Nanobot open source?: Yes. Nanobot is open source under the MIT licence. Its source code is on GitHub at HKUDS/nanobot and is written mainly in Python.
Is Nanobot free?: Yes. Nanobot is open source, so the software itself is free to use.
Can I self-host Nanobot?: Yes. Nanobot can be self-hosted on your own server or infrastructure; there is no official hosted version.
What are some alternatives to Nanobot?: Similar open-source tools in the AI Tools category include QwenPaw, GPT Researcher and Poco Claw. SaaS products in the same category include Aera Browser, August and CalmPilot.
Is Nanobot actively maintained?: Yes. The most recent commit to Nanobot was on 2 October 2026, and the latest release is v0.3.5, published on 15 September 2026. The project has 49k stars on GitHub.