> ## Documentation Index
> Fetch the complete documentation index at: https://www.getmaxim.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# AI-powered multi-agent support system

> Multi-agent systems are a powerful way to build complex applications that can handle a wide variety of tasks.

export const MaximPlayer = ({url}) => {
  return <iframe className="border-background-highlight-secondary h-full w-full rounded-md border-2 aspect-video" src={url} allow="accelerometer; autoplay; encrypted-media; gyroscope; picture-in-picture" allowFullScreen></iframe>;
};

## How to create a multi-agent system?

Create an intelligent workflow that can automatically understand and answer user queries related to their accounts, billings, returns or any FAQ.

<img src="https://mintcdn.com/maximai/SwDaTi397y_biFsK/images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system.png?fit=max&auto=format&n=SwDaTi397y_biFsK&q=85&s=53bcb8913cc2d25f18fbe3aa8c102005" alt="Multi agent system workflow" width="2894" height="1626" data-path="images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system.png" />

<Steps>
  <Step title="Set up user query classification agent">
    The below prompt would act as our routing agent, classifying the user query into different categories (Account, Billing, Returns & FAQ).

    <img src="https://mintcdn.com/maximai/SwDaTi397y_biFsK/images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system-routing-agent.png?fit=max&auto=format&n=SwDaTi397y_biFsK&q=85&s=0c5840ab053ab15595291d2805376986" alt="Multi agent system routing agent" width="816" height="1416" data-path="images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system-routing-agent.png" />
  </Step>

  <Step title="Route data based on category">
    Use the code block as a helper to route the data to the corresponding agent based on the category.

    <img src="https://mintcdn.com/maximai/SwDaTi397y_biFsK/images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system-code.png?fit=max&auto=format&n=SwDaTi397y_biFsK&q=85&s=8a034f374253b04bb02e7c23b4cb9716" alt="Multi agent system routing helper block" width="720" height="1536" data-path="images/docs/evaluate/how-to/prompt-chains/tutorials/multi-agent-system-code.png" />
  </Step>

  <Step title="Create the required sub-agents">
    Use the [Prompt](/prompt-engineering/prompt-playground) feature to create the required individual sub-agents. Make sure to attach the required tools for every sub-agent to process the user query.

    Additionally, connect the outputs of all the agents to the `Final output node` to get the final response.
  </Step>

  <Step title="Test the agent">
    Here is a video of the agent in action:

    <MaximPlayer url="https://drive.google.com/file/d/1-Cg1LUgVrUWVojKw_K3_CXfP_z70-TWf/preview" />
  </Step>
</Steps>