How GrokBot Coordinates Teams of AI Agents

Riley Brown20:03
0 comments · 0 votesOpen discussion
Sign in to join the discussion

    Video summary

    Riley Brown demonstrates creating separate GrokBot agents for email, content and development work. Each agent gets a name, role description and its own cloud computer, while skills and plugins are shared across the workspace and conversations stay synchronized between desktop and mobile.

    The video follows a development agent asking a content agent for transcript themes, choosing an application idea and building the result in its cloud environment. Their agent-to-agent messages remain visible, making the handoff inspectable rather than hiding the collaboration behind one combined answer.

    GrokBot also supports scheduled routines and event triggers from services such as Slack, Git, Linear, Sentry and PagerDuty. Brown notes current gaps, including limited trigger coverage and unfinished file and group features, but argues that persistent agents with computers, routines and direct delegation form a clearer operating model for multi-agent work.

    Original YouTube thumbnailWatch on YouTube