App Agent

A Loop AI agent is a specialist for one business function, loaded with that function’s skills, knowledge and frameworks. It is not a general-purpose chatbot. Each agent knows which deliverables it produces, which questions to ask before starting, and when to hand work to another specialist.

Loop agents come with role-specific methods (such as MEDDPICC for sales or Dunford positioning for PMM) and read the content you keep in Loop: brand guidelines, CRM data and past deliverables. Outputs are based on your company’s context and not stock examples. They also know when to route work to another specialist and pass along the context.

Open a chat with the agent and describe the task in plain language: what you need, who it is for and any context that matters. Most agents load related content from Loop when you @mention it. If the agent needs more before it can produce something useful, it asks one focused question and not a long intake form.

Your content, brand guidelines, CRM data and deliverables are scoped to your company and never shared with other companies on Loop. Agents read your data to produce relevant output. They do not change external systems such as your CRM unless you take an explicit action.

Yes. Every agent works from your company’s brand voice and the content you keep in Loop. @mention your own content, templates and brand guidelines and the agent’s output follows your team’s format and context. Creating your own agents is on the Enterprise plan.