[
    {
        "prompt1": "A Snap in my pipeline is failing with a 'NullPointerException' error. What steps do I take to identify and resolve the issue?",
        "prompt2": "My pipeline is not processing records as expected, and the Snap logs show an 'Invalid Query' error. What steps can I take to diagnose and fix the query issue?"
    }
]

The user is asking about SnapLogic's features. Provide a concise, helpful response. 
Question: {{prompt1}} Answer:
Question: {{prompt2}} Answer:

On validation, the Snap replaces the placeholders with actual values, and displays the generated prompt.

Amazon Bedrock Prompt Generator Snap configuration	Amazon Bedrock Prompt Generator Snap output

These settings allow you to fine-tune the message response, ensuring the generated text matches your desired conversational style and tone. In this example, the System prompt is set to: You are a SnapLogic Support Executive. Provide concise, accurate answers to SnapLogic product-related questions. to ensure the specified model assumes this role while generating the responses. The Stop sequences is set to: ["Thank you for your question.", "If you have any more questions, feel free to ask.", "<|endofresponse|>"] to ensure that the model stops generating text once it reaches this instruction, providing a precise and relevant response for each FAQ question.

On validation, the Snap displays the generated FAQ responses to the two questions separately, with the usage details.

Amazon Bedrock Converse API Snap configuration	Amazon Bedrock Converse API Snap output