HKUDS/nanobot

This page introduces an external open-source repository. It is not an HDATF product.

What it does

nanobot is a self-hosted personal AI agent framework in Python that runs in a WebUI, a terminal or chat apps. It combines tools, long-term memory, MCP, subagents, scheduled automations and an OpenAI-compatible API.

How it helps ATF

Its small core that combines tools, long-term memory, subagents and scheduled runs could be a reference for Harness. Its connections to chat apps such as Slack and email are worth comparing with the messaging features of ATF Works.

License

MIT Permissive. Commercial use and changes are allowed if the copyright notice is kept.

More in this category

• rowboatlabs/rowboat
  An open-source AI coworker where each teammate runs their own agent on their own machine, with memory of email, meetings, notes and code. Shared Spaces hold messages, a markdown wiki, files and whiteboards where these agents post results.
• nanocoai/nanoclaw
  NanoClaw is a lightweight alternative to OpenClaw that runs each agent in its own Linux container on Anthropic's Agents SDK. It connects to Slack, Telegram, Discord, WhatsApp, Gmail and other apps, with memory and scheduled jobs.
• danielmiessler/LifeOS
  LifeOS is a general purpose AI harness that captures a person's values and goals and uses that context to help with life and work tasks. It centers on moving from a current state to an ideal state.
• cloudflare/moltworker
  An experimental proof of concept that runs the OpenClaw personal AI assistant in a Cloudflare Sandbox container, using Cloudflare Access for authentication, Browser Rendering for browsing, and optional R2 storage for persistence.
• sipeed/picoclaw
  PicoClaw is a lightweight personal AI assistant written in Go and inspired by NanoBot, built to run on low-cost hardware with little memory. The project says it is in early development and not ready for production use.

Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.