Amazon Q in Connect do not support multi turn conversation by Agent manual input

Amazon Q in Connect does support multi-turn conversations. The agentic Retrieval Augmented Generation (RAG) capabilities in Amazon Q enable it to maintain conversational context across interactions by storing short-term memory, which allows for natural follow-up questions without requiring users to restate previous context.

When agents interact with Amazon Q in Connect, the system is designed to maintain context throughout the conversation. If you're experiencing an issue where follow-up questions are being treated as new topics, there are a few potential approaches to address this:

1. Ensure you're using the latest version of Amazon Q in Connect, as the multi-turn conversation capabilities are part of its core functionality.

2. You may need to customize the AI Prompts and AI Agents that control Amazon Q's behavior. This can be done through the Amazon Q in Connect APIs, where you can create custom AI Prompts that specifically instruct the system to maintain context between questions.

3. Check your configuration for Amazon Q in Connect, particularly the AI Agent settings. You can create an AI Agent using the CreateAIAgent API and configure it with appropriate AI Prompts that enhance the conversational capabilities.

4. If you're using Amazon Q in Connect with calls, verify that Contact Lens conversational analytics is enabled, as this is required for call support (though not for chat or email).

The system should be able to handle semantic ambiguity and maintain both the conversation state and previously retrieved contextual data in-memory, enabling more accurate and relevant responses to follow-up questions.
Sources
Bringing agentic Retrieval Augmented Generation to Amazon Q Business | Artificial Intelligence
Personalize agent assistance with Amazon Q in Connect | AWS Contact Center
Use Amazon Q in Connect for generative AI–powered agent assistance in real-time - Amazon Connect