Agor, multiplayer AI – stop running agents alone (github.com)

Agor has introduced a groundbreaking platform designed for collaborative AI development, allowing teams to work together on live spatial boards. These boards facilitate real-time interactions among team members and AI agents, enabling collaborative coding, knowledge sharing, and integrated workflows. Users can raise AI teammates that retain memory and context, enhancing collaboration and reducing reliance on individual setups. Agor operates without a proprietary model, allowing installation of coding agents like Claude Code and Codex, which can be run seamlessly in isolated environments, thus avoiding vendor lock-in. This innovation is significant for the AI/ML community as it transforms how teams interact with AI by making the process collectively accessible. Agor enables simultaneous viewing and participation in various coding tasks, streamlining communication through live cursors, comments, and session tracking. Additionally, the platform supports multi-runtime capabilities and structured environments for diverse coding agents. With features like shared knowledge bases, cron-style scheduling for workflows, and customizable dev environments, Agor empowers teams to maximize their efficiency and improve AI training and application through collaborative efforts. This shift towards multiplayer AI represents a pivotal evolution in managing and maximizing AI's potential in team settings.