multipeer is built around a simple idea: AI agents should not be a solitary experience. Rather than keeping each agent session locked inside one developer’s terminal, the app places agents, terminals, files, and notes on a shared infinite canvas that multiple people can view and work on at the same time. The tagline, running your agents in multiplayer, captures the intent: turning agent work into something a team does together, live.
The core of the product is that canvas. Agent sessions from Claude Code, Codex, and Gemini can run side by side, each showing its own status, such as working, needs you, or done, so teammates can see at a glance what is running and what needs attention. Terminals, project files, and notes live in the same space, which means context stays visible instead of being scattered across chat threads and local folders. Sharing over the internet is part of the design, so collaborators do not need to be in the same room.
It is aimed primarily at developers and small product teams who already use command-line coding agents and want to coordinate around them. That includes pairs reviewing an agent’s output together, teams splitting work across several agents at once, and anyone who finds solo agent sessions hard to track. Because the app is macOS-first, the current audience is Mac-based developers, with Windows, Linux, Android, and iPhone versions listed as planned.
Access is currently through a waitlist rather than an open download. Visitors pick their pl
Comments & Ratings
0No comments yet. Be the first to review.