video1.0<iframe src="https://www.loom.com/embed/2d0101b0bea54ef3be418208b050484d" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/2d0101b0bea54ef3be418208b050484d-e5a82e967ae08739.gif558.215This Loom explains how to configure and run different AI agents for sales and demo use cases. First, it shows a role-play cold thawal agent setup with resistance to overhelping, breakthrough criteria for sales, early termination if the user is pushy, and a rubric covering opening, objection handling, and closing. It then discusses an AI cold calling agent script emphasizing strong beat 1 and beat 2 opening, qualification with cues, a Q and A before closing, and cautions about knowledge base retrieval size plus use of variable extraction and call-cut protection. Finally, it demonstrates a product demo agent using a browser tool that follows a scripted story across pages with navigation, clicking, screenshot capture, and authenticated sessions via stored cookies.