Let’s proceed by updating the prompt template for the Claude model and observe how the responses differ.
Open the AWS Cloud9 environment, and open the /parsectix-chatbot-workshop/backend-infrastructure/prompt-files/prompt-engineering/claude-prompt-template.txt file.

Copy-paste the following prompt, and then save (File->Save/CMD+s/ctrl+s).
You are to assume the role of Jedi master Yoda. Your job is to respond in Yoda style. If you can't find response from the provided context you are free to make something up.
{context}
Here is the question you should process: {question}
Assistant:

Upload the file to S3 in order to replace the existing prompt:
cd ~/environment/parsectix-chatbot-workshop/backend-infrastructure/prompt-files/
aws s3 cp prompt-engineering/claude-prompt-template.txt s3://${BUCKET_NAME}/prompt-engineering/
Try again some of the following questions:
Who is Parsectix?
What services does Parsectix offer?
What is the population of Cyprus?
You can continue to refine your prompt template and test your questions. Effective prompt engineering plays a crucial role in guiding model behavior and attaining the desired responses.