Extract XML from llm response

Function: Extract XML from LLM Response

This action helps you automatically find and pull out XML text from a longer response generated by an AI (Large Language Model). This is useful when an AI provides information embedded within a larger message, and you only need the structured XML part.

Input

• LLM Response (Text): The full text response from your AI model that you expect to contain XML. This input is required.

Output

• result (Text): The extracted XML text. If the action successfully finds XML, it will store it in a variable named "RESULT" by default, or any other name you specify. If no XML is found, this variable will not be updated by this action.

Execution Flow

Real-Life Examples

Here are some practical examples of how you can use the "Extract XML from LLM Response" action:

Example 1: AI Provides Pure XML Configuration

Scenario: You ask an AI to generate a configuration file in XML format, and its response is solely the XML content.

• Inputs:
  • LLM Response:

    <configuration>
        <setting name="theme" value="dark"/>
        <setting name="notifications" value="enabled"/>
    </configuration>
• Result: The variable RESULT will contain:

  <configuration>
      <setting name="theme" value="dark"/>
      <setting name="notifications" value="enabled"/>
  </configuration>

Example 2: AI Provides XML Embedded in a Conversational Message

Scenario: You ask an AI for a product description, and it provides some introductory text followed by the product details in an XML block.

• Inputs:

  • LLM Response:

    "Here is the product information you requested:
    
    ```xml
    <product>
        <id>P123</id>
        <name>Wireless Earbuds</name>
        <price currency="USD">99.99</price>
        <features>
            <feature>Noise Cancelling</feature>
            <feature>Bluetooth 5.0</feature>
        </features>
    </product>

    Let me know if you need anything else!"

• Result: The variable RESULT will contain:

  <product>
      <id>P123</id>
      <name>Wireless Earbuds</name>
      <price currency="USD">99.99</price>
      <features>
          <feature>Noise Cancelling</feature>
          <feature>Bluetooth 5.0</feature>
      </features>
  </product>

Example 3: AI Provides Text Without Recognizable XML

Scenario: You ask an AI a question, and its response is plain text without any XML structure or XML code blocks.

• Inputs:
  • LLM Response:

    "The capital of France is Paris. It is known for its art, fashion, gastronomy, and culture."
• Result: The variable RESULT will not be updated by this action, as no XML was found. You might want to add a check in your workflow to see if RESULT is empty after this action to handle cases where no XML is extracted.