Thank you for your feedback, Aalap. I appreciate the effort you put into testing this approach.

You are absolutely right that LLMs do not inherently recognize symbols like +++ unless they are properly defined. Prompt Decorators are not meant to be magic switches but structured directives that guide AI responses—when they are first introduced and defined correctly. If they were tested without defining them in the system prompt or context, the model would indeed treat them as regular text, which might explain your results.

To clarify, I have shared a structured implementation of this approach in the GitHub repository, which is also mentioned in the post:

🔗 https://github.com/smkalami/prompt-decorators

That being said, it is important to note that this is not an ideal implementation, and others may find better ways to define and use these Prompt Decorators. The current approach is simply a realization of the idea, a way to enable users to define structured directives and reuse them efficiently in their daily AI interactions.

I welcome further discussions on refining this method, and I would love to hear more about your testing process. Looking forward to your insights!