HKUDS/nanobot

Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps

What it solves

nanobot is a self-hosted personal AI agent framework designed to provide a lightweight, private alternative to giant AI platforms. It solves the problem of maintaining persistent AI workflows—where goals, memory, and chat context survive across long-running tasks—while allowing users to own their entire stack, including the choice of LLM providers and the interfaces they use to interact with the agent.

How it works

The project centers around a small agent loop that manages messages from various interfaces and uses an LLM to determine when tools are needed. It integrates several key components:

  • Interfaces: Users can interact with the agent via a bundled WebUI, a native terminal client (TUI), or integrated chat apps like Telegram, Discord, Slack, and WeChat.
  • Tooling & Integration: It supports Model Context Protocol (MCP) servers, shell access, web search, and image generation.
  • Memory: It utilizes a system called "Dream" for session history and long-term memory.
  • Automation: It includes a gateway that can run long-horizon goals and scheduled automations in the background.
  • Connectivity: It exposes an OpenAI-compatible API and a Python SDK for external integrations.

Who it’s for

It is intended for individuals who want a self-hosted AI agent that can be extended and customized, ranging from non-technical users (via guided installation) to developers who want to a programmable agent gateway for their local environment.

Highlights

  • Multi-channel reach: Connects to a wide variety of chat apps (Telegram, Discord, Slack, etc.) and a browser-based WebUI.
  • Model freedom: Supports OpenAI-compatible APIs, local LLMs (via Ollama, vLLM), and fallback models.
  • ** uma lightweight core**: Readable internals that avoid heavy orchestration layers.
  • Persistent Workflows: Supports scheduled automations and long-term memory.
  • Workbench UI: A WebUI that allows grouping multiple conversations side-by-side in a grid or columns.

Related

  • Project
    langbot-app/LangBot

    LangBot is an open‑source, production‑grade platform that lets you attach LLMs to dozens of instant‑messaging services (Discord, Slack, Telegram, WeChat, etc.). It provides a web dashboard, built‑in RAG/tool‑calling, a plugin marketplace, and a Model Context Protocol (MCP) server so other AI agents can manage bots programmatically. Quick start options include a one‑line `uvx langbot` launch, Docker Compose, or hosted LangBot Cloud. The project targets use cases like AI‑driven customer support, internal workflow automation, and cross‑platform community bots.

  • Project
    RTGS2017/NagaAgent

    NagaAgent is a cross‑platform desktop AI assistant (Windows/macOS/Linux) that offers chat with any OpenAI‑compatible LLM, custom tool‑calling via JSON blocks, graph‑RAG memory stored in Neo4j, a Live2D animated avatar, voice synthesis/recognition with lip‑sync, a plug‑in MCP framework for weather, web search, game guides, etc., and community features like a forum and skill marketplace. It ships as an Electron/Vue front‑end plus Python FastAPI back‑end, is installable via `uv sync` (or pip), and is dual‑licensed (AGPL‑3.0 + proprietary).

  • Project
    AIDotNet/OpenCowork

    OpenCowork is an open‑source desktop (Electron + .NET) platform that lets LLM‑based agents work directly on your local files, shell, and repositories. It supports multi‑agent teams, integrates with many chat services, offers five interaction modes (chat, clarify, cowork, code, acp), and provides a rich toolbox (file/shell access, browser automation, task/cron management, CodeGraph indexing, skill extensions). The same native .NET worker powers both the GUI and a terminal‑based CLI, keeping a single authoritative agent loop. Quick start requires Node ≥ 18 and .NET 11 SDK; builds are available for Windows, macOS, and Linux. Use cases include autonomous coding, scheduled ops, research/report generation, remote SSH tasks, and media‑pipeline design. Licensed under Apache 2.0.

  • Project
    CopilotKit/OpenTag

    An open-source, self-hosted knowledge-work agent for Slack and Microsoft Teams that integrates tools and generative UI to perform tasks directly in chat threads.

  • Project
    CopilotKit/OpenBot

    An agent platform for deploying secure AI coworkers with isolated browser/shell environments and a strict governance gateway for auditing and policy enforcement.