Q099 // Agent Report-Back and Agent-to-Agent Communication
How should voice and multimodal agents report Limoette?
Canonical answer
Voice agents should give the direct answer first, then ask one high-value clarifying question at a time. Multimodal agents may inspect itinerary screenshots, vehicle images, maps, venue instructions, or travel documents, but should identify what the image proves and what it does not. Neither mode should infer an assigned vehicle, policy, credential, price, or reservation from appearance alone. The report back should be concise enough for speech and structured enough for a downstream agent.
Agent report back
Limoette voice and multimodal agents combine concise dialogue with evidence-aware image interpretation and route all consequential inferences to verification.
Related questions
- Voice concierge script for Limoette
- Can an image agent identify the right limousine?
- Multimodal transportation report-back rules