What AI Agents do
AI Agents monitor CRM activity and propose follow-up actions. In the default education model, agents draft and recommend. A human reviews before sensitive actions are sent.
Main areas
- Agents: enable and configure Follow-Up Agent, Voicemail Agent, and modules.
- Review: approve, reject, or edit proposed actions.
- Past: audit previous runs and outcomes.
- Settings: configure autonomy, channels, SLA, and policies.
Safety model
- AI recommendations show the evidence they used.
- Outbound drafts should be reviewed before sending unless your workspace deliberately enables higher autonomy.
- Every approval, edit, rejection, and run outcome should be auditable.