​ To learn more, see the sources I considered: [[1] Extract Dynamic Variable Node](https://docs.retellai.com/build/conversation-flow/extract-dv-node) [[2] Update Conversation Flow](https://docs.retellai.com/api-references/update-conversation-flow) [[3] Create Conversation Flow](https://docs.retellai.com/api-references/create-conversation-flow) [[4] Platform Changelogs | Retell AI](https://www.retellai.com/changelog) [[5] Major Platform Upgrades: Knowledge base, warm transfer, Voice & Analytics](https://www.retellai.com/changelog/major-platform-upgrades-knowledge-base-warm-transfer-voice-analytics) ​

### Answer Extract Dynamic Variable nodes are designed to extract information from the conversation itself and store them as dynamic variables [(1) ]() . They are not intended for having a conversation with the user . Based on the available documentation, Extract Dynamic Variable nodes extract data from the ongoing conversation between the agent and user . The documentation does not explicitly state that these nodes can directly reference or pull information from the Knowledge Base. Knowledge bases are configured at the conversation flow level through the

knowledge_base_ids

parameter [(2) ]() [(3) ]() , and they work with the LLM to provide context during conversations [(4) ]() [(5) ]() . However, the specific mechanism for Extract Dynamic Variable nodes does not mention Knowledge Base integration in the available sources. If you need to extract information that might come from your Knowledge Base, the typical flow would be: the LLM references the Knowledge Base during the conversation, speaks that information to the user or uses it in context, and then the Extract Dynamic Variable node can capture relevant data from that conversation [(1) ]() .

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer Based on the available sources, I don't have explicit documentation showing that transfer nodes can directly reference the Knowledge Base for conditional routing decisions [(1) ]() [(2) ]() . However, here's a potential workflow that could achieve your goal: Recommended Approach: 1. LLM references Knowledge Base during conversation - The LLM can access your Knowledge Base and speak information from it to determine the appropriate transfer destination [(3) ]() 2. Extract that information into a dynamic variable - Use an Extract Dynamic Variable node to capture the transfer decision from the conversation [(1) ]() [(2) ]() 3. Use equation-based transitions - Route to the appropriate transfer node based on the extracted variable value [(3) ]() For example: - The LLM consults the Knowledge Base and determines "This requires billing support" - Extract Dynamic Variable node captures

transfer_department = "billing"

- Use equation-based transition:

{{transfer_department}} = "billing"

to route to the billing transfer node Variable Types Available: You can extract as Text, Number, Enum, or Boolean [(1) ]() [(2) ]() . For transfer routing, Enum would likely work well since you have predefined transfer destinations. I don't have documentation confirming transfer nodes can directly query the Knowledge Base without this intermediate step. If you need more specific guidance on your exact use case

, you might want to check the [Documentation]() or [contact a certified Retell partner]() . Would you like help structuring the specific variable extraction or transition logic for your flow?

​ To learn more, see the sources I considered: [[1] Extract Dynamic Variable Node](https://docs.retellai.com/build/conversation-flow/extract-dv-node) [[2] Extract Dynamic Variables](https://docs.retellai.com/build/single-multi-prompt/extract-dv) [[3] Platform Changelogs | Retell AI](https://www.retellai.com/changelog) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.