### Answer I don't have information in the available sources that specifically explains the "Store Call Info as Variables (Optional) JSON" field on multi-prompt agent edges. The sources cover multi-prompt agent architecture, state transitions, and context preservation [(1) ]() , but don't detail this specific JSON configuration option. However, based on the general multi-prompt architecture, this feature likely allows you to capture and store conversation data as it flows between states, enabling context preservation across state transitions . Check out these potentially useful sources: - [Single/Multi Prompt Agent Overview]() - [Function Calling Overview]() Would you like me to provide a general answer based on my knowledge of how variable storage typically works in conversational AI systems? For more specific guidance on this feature, I'd recommend visiting [Contact Sales]() or [) .

​ To learn more, see the sources I considered: [[1] Single/Multi Prompt Agent Overview](https://docs.retellai.com/build/single-multi-prompt/prompt-overview) ​

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

@AI Evy can you add additional function or command within the edge?

### Answer Based on the available sources, I don't have specific information about adding functions or commands directly within edges for multi-prompt agents. For multi-prompt agents, functions are typically added at two levels: [(1) ]() - Agent level – general functions that can be called multiple times across prompts (like end call) - State level – functions limited to that specific state For Conversation Flow agents, edges connect nodes and define transition conditions, but functions are executed through dedicated Function Nodes rather than within the edges themselves. [(2) ]() [(3) ]() Check out these potentially useful sources: - [Integrate any system with custom function]() - [Function Node Overview]() Would you like me to provide a general answer, or would you like more details about how to structure functions within states for your multi-prompt agent?

​ To learn more, see the sources I considered: [[1] Discord message from @simonyost on 2025-01-24](https://discord.com/channels/1186424376300679209/1294891368505937952/1332388372362035271) [[2] Function Node Overview](https://docs.retellai.com/build/conversation-flow/function-node) [[3] Retell AI’s Advanced Conversation Flow | Retell AI](https://www.retellai.com/blog/unlocking-complex-interactions-with-retell-ais-conversation-flow) ​

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

### Answer ### Answer To prevent your agent from mixing English when speaking Spanish, you need to address two things: 1. Set the language setting to Spanish The language setting determines what language the agent understands (speech-to-text) and influences pronunciation. [(1) ]() You can set this via the dashboard or API using the

language

field (e.g.,

es-ES

for Spain Spanish or

es-419

for Latin American Spanish) . [(2) ]() 2. Explicitly instruct the LLM to respond only in Spanish The language setting does not enforce the agent to generate responses in a certain language—you have to set that in the prompt yourself. [(1) ]() Add something like

always respond in Spanish

into your prompt. A community member resolved a similar issue (agent speaking English despite selecting Japanese) by using this prompt structure: [(3) ]()

# Role 
You are a Spanish-speaking agent working at [your company]

# Response Guidelines
 - You can only speak in Spanish

This combination of setting the correct language code and adding explicit prompt instructions should prevent the English mixing.

​ To learn more, see the sources I considered: [[1] Set language for your agent](https://docs.retellai.com/agent/language) [[2] Update Voice Agent](https://docs.retellai.com/api-references/update-agent) [[3] Language problem](https://discord.com/channels/1186424376300679209/1421806667343073411) ​

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