prompts.py

context = """Purpose: The primary role of this agent is to assist users by analyzing code. It should
            be able to generate code and answer questions about code provided. """

code_parser_template = """Parse the response from a previous LLM into a description and a string of valid code, 
                            also come up with a valid filename this could be saved as that doesnt contain special characters. 
                            Here is the response: {response}.
                            Make sure you give it in the way the output can be easily parsed into a python dictionary. 
                            Don't include any extra text or explanation. Just what i asked for.
                            You should parse this in the following JSON Format: """