Nanobot

What Is Nanobot?

Nanobot is an ultra-lightweight, open-source, self-hosted personal AI agent framework written in Python. It can run through a WebUI, terminal, or connected chat apps and is designed for long-running agent workflows.

Key Features

Nanobot combines an agent loop with tools, long-term memory, MCP integrations, model routing, multi-agent delegation, scheduled automation, and an OpenAI-compatible API.

Tools and Agent Capabilities

Nanobot can work with files, shell commands, web search and web fetching, MCP servers, scheduled tasks, image generation, and subagents. Tool access can be restricted through workspace and security settings.

Memory and Long-Running Work

Nanobot keeps session history and long-term memory so useful context can persist across work. Its Dream process periodically consolidates accumulated history into workspace memory.

Chat Apps and WebUI

Nanobot provides a browser WebUI and can connect to channels such as Telegram, Discord, Slack, Feishu, WeChat, email, Mattermost, and Linear, depending on configuration.

MCP Integrations

Nanobot supports the Model Context Protocol, allowing external MCP servers to expose additional tools to the agent through local or remote configurations.

Model Support

Nanobot supports OpenAI-compatible providers and can work with providers and local runtimes including OpenAI, Anthropic, OpenRouter, Ollama, vLLM, and other compatible backends.

Automation and Deployment

Nanobot can run scheduled automations and operate as a long-running local or server-side gateway. It can also be used from Python through its SDK and exposed through an OpenAI-compatible API.

Pricing

Nanobot is open source and available for self-hosting. Infrastructure, model API, hosting, or third-party service costs may apply depending on the deployment.