GoKawiilNanoMuse launches as open-source personal AI agent for phone, desktop and web
Developers have released nanoMuse, an open-source AI agent licensed under GPL-3.0-or-later that runs across Android, iOS, macOS, Windows and Linux, plus a web console and self-hostable relay. The project, styled after Meta's Muse, performs tasks autonomously, keeps running when the app is closed, retains memory of the user, and pauses to confirm before irreversible actions. Users get a free allowance of model use on a community relay subsidized by the developer, or can supply their own API key or self-host the relay.
GoKawiil's interpretation of the reporting above, not reported fact.
By open-sourcing the full stack — apps, web console and relay — the project invites community self-hosting, which could appeal to privacy-conscious users wary of cloud-based AI agents. Its free community relay, funded by the developer, suggests a non-profit, donation-reliant model that may face sustainability questions as usage scales. The agent's action-taking, cross-device design places it in a growing category of 'agentic' AI tools competing with offerings from larger tech companies.
- nanoMuse is fully open source (GPL-3.0-or-later) and covers phone, desktop, web, and relay components.
- It performs autonomous tasks, persists after the app closes, and asks confirmation before irreversible actions.
- Users can rely on a free community relay subsidized by the developer, bring their own API key, or self-host via Docker.
Source: github.com, 2026-10-07
Published there as: “Show HN: NanoMuse – An open-source AI agent for your phone and computer”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.