​ To learn more, see the sources I considered: [[1] List Retell LLMs](https://docs.retellai.com/api-references/list-retell-llms) [[2] Get Retell LLM](https://docs.retellai.com/api-references/get-retell-llm) [[3] Create Conversation Flow](https://docs.retellai.com/api-references/create-conversation-flow) [[4] Update Conversation Flow](https://docs.retellai.com/api-references/update-conversation-flow) [[5] Prompt Engineering Guide](https://docs.retellai.com/build/prompt-engineering-guide) ​

### Answer The global prompt in Retell AI is a prompt that applies across the entire conversation, regardless of which state or node the agent is in. [(1) ]() [(2) ]() ## Where Global Prompts Are Used For Single/Multi-Prompt Agents: - The

general_prompt

field contains the global prompt appended to the system prompt no matter what state the agent is in. [(2) ]() For Conversation Flow Agents: - The

global_prompt

field is used in every node of the conversation flow. [(3) ]() [(4) ]() ## How to Set Global Prompts Via API for Conversation Flow:

javascript
const conversationFlowResponse = await client.conversationFlow.create({
  model_choice: { model: 'gpt-4.1', type: 'cascading' },
  global_prompt: "You are a helpful customer service agent.",
  nodes: [
    {
      id: 'start',
      instruction: { text: 'Greet the customer and ask how you can help them.', type: 'prompt' },
      type: 'conversation',
    },
  ],
  start_speaker: 'agent',
}) ;

[(3) ]() ## Best Practice: Use Sectional Prompts Break large global prompts into focused sections for better organization: [(5) ]()

///## Identity
You are a friendly AI assistant for [Company Name].

///## Style Guardrails
Be concise: Keep responses under 2 sentences unless explaining complex topics.
Be conversational: Use natural language and contractions.

///## Response Guidelines
Return dates in spoken form: Say "January fifteenth" not "1/15".
Ask one question at a time.

This structured approach improves reusability, maintainability, and LLM comprehension.

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