Extract XML from llm response

Function: Extract XML from AI Assistant's Reply

This action helps you automatically find and extract XML formatted text from a longer response provided by an AI assistant (Large Language Model - LLM). This is useful when an AI assistant generates a response that includes structured data in XML format, along with other conversational text.

Input

• LLM Response (A piece of text): This is the full text generated by the AI assistant, which might contain the XML you want to extract. This input is required.

Output

• Result (A piece of text): This is the variable where the extracted XML text will be stored. By default, it will be named "RESULT", but you can choose a different name for your variable.

Execution Flow

Real-Life Examples

Here are some examples of how you can use the "Extract XML from AI Assistant's Reply" action:

Example 1: Extracting Product Details from a Chatbot Response

Imagine you're building an e-commerce support chatbot. A customer asks for details about a product, and your AI assistant responds with a mix of conversational text and structured product information.

• Inputs:
  • LLM Response: "Certainly! Here are the details for the 'Smartwatch Pro': xml<product><name>Smartwatch Pro</name><sku>SWP-2023</sku><price>199.99</price><features><feature>Heart Rate Monitor</feature><feature>GPS</feature><feature>Waterproof</feature></features></product> Let me know if you need anything else!"
• Result: The action will extract the XML block and store it in a variable named ProductDetailsXML.
  • ProductDetailsXML will contain:

    <product>
        <name>Smartwatch Pro</name>
        <sku>SWP-2023</sku>
        <price>199.99</price>
        <features>
            <feature>Heart Rate Monitor</feature>
            <feature>GPS</feature>
            <feature>Waterproof</feature>
        </features>
    </product>

Example 2: Processing a Configuration Update from an AI Assistant

You're using an AI assistant to help manage system configurations. You ask it to generate a configuration snippet, and it provides it within a larger explanation.

• Inputs:
  • LLM Response: "I've generated the requested configuration for the new user role. Please review and apply it carefully. xml<config><role name='Admin'><permissions><permission>read</permission><permission>write</permission></permissions></role></config> This should grant the necessary access."
• Result: The action will find and store the configuration XML in a variable named NewRoleConfig.
  • NewRoleConfig will contain:

    <config>
        <role name='Admin'>
            <permissions>
                <permission>read</permission>
                <permission>write</permission>
            </permissions>
        </role>
    </config>

Example 3: Extracting a Data Report from an AI-Generated Summary

Your AI assistant summarizes daily sales data and includes a detailed XML report for further processing.

• Inputs:
  • LLM Response: "Daily sales summary: Overall, sales were up by 5% today. Here is the detailed breakdown: xml<report date='2023-10-27'><totalSales>15000</totalSales><itemsSold>300</itemsSold><topProduct>Widget X</topProduct></report> We should focus on promoting Widget Y next."
• Result: The action will extract the report XML and save it to a variable named DailySalesReport.
  • DailySalesReport will contain:

    <report date='2023-10-27'>
        <totalSales>15000</totalSales>
        <itemsSold>300</itemsSold>
        <topProduct>Widget X</topProduct>
    </report>