Prompt Template and Documentation Clarification

[DoTrONoL]

Summary:

I'm experiencing confusion with your plugin's prompt template and it's documentation, and because of that I'd like to clarify some points to ensure I understand it correctly and also suggest it to be revised. FYI, I'm a newcomer in the AI field, so I know almost nothing about it.

• Permalink

Questions and Concerns:

1. Definition of "task" and "instructions":
   • Can you please clarify what "# Task" represents in the context of the prompt template? It seems to be related to selected text, but it's mentioned differently in the "Caveats" section.
   • Are "instructions" equivalent to "chat prompts"?
2. Use of "#" symbol:
   • Is the "#" symbol used to impact the prompt or is it only used for documentation purposes?
   • I've noticed that in the "Enhanced Actions (Context)" section, there's an inconsistency in naming conventions (i.e., "context:" and "# Relevant context"; are they interchangeable?).
3. Prompt Order:
   • How would the final order of the prompt look like? I imagine it would be something like:
     • System Prompt
     • Chat Prompt
     • Selected Text
     • Context
   • However, in your example, it seems to be:
     • System Prompt
     • Task? (what is this, is it somehow related to the actions in the plugin's settings?)
     • Chat Prompt (is this equivalent to instructions?)
     • Selected Text
     • Context
   • I think this order would be better for the plugin; please consider it as a suggestion. I'm unsure about the implications of this change, and I would be grateful if you do answer.
     • System Prompt
     • Chat Prompt
     • Context
     • Selected Text
4. Clarification on "context":
   • I assume that the term "context" is used to refer to an external document that it's inside a [[]] of a selected text, but it's definition and usage are not explicitly stated in the documentation.
5. Documentation Clarity:
   • The documentation assumes prior knowledge of AI concepts, which is not my case as a beginner.
   • Can you please simplify the language and provide clearer explanations for those new to AI?
6. Placeholder Usage:
   • While I think I understand how to use placeholders correctly, but the documentation only confuses me further, mainly in the "Caveats" section. The reasons were mentioned in the previous items.

Additional Feedback:

If you'd like more detailed explanations on specific points, please let me know.