Starnet is a local-first desktop application that visualizes AI agents as a living pixel-art space station crew. Instead of watching terminal logs scroll, you see your agents as characters moving around a station while they do real work through an agent harness – reading files, running tools and completing tasks with models you connect using your own API keys. It is an open-source project on GitHub with around 1,600 stars, actively developed in 2026.
Agent tooling is usually invisible: long-running work happens in a terminal tab and status is hard to read at a glance. Starnet wraps a multi-agent harness in a game-like Tauri desktop interface, where each agent has a visible presence and activity, making parallel agent work legible and, frankly, fun to supervise. It is local-first and bring-your-own-key, with OpenRouter among the supported providers, so data and credentials stay on your machine while the crew runs. The project blends a playful presentation with a real execution layer underneath.
Starnet is free and open source under the MIT license. You only pay your own model providers for API usage.
Developers and agent enthusiasts who run local AI agents and want an engaging, visual way to supervise their work.