nite f1214be148 feat(llm-integration): Enhance prompt clarity and unify PDF attachment ...

This commit improves the structure and clarity of the prompt sent to the LLM (Gemini/OpenAI) in the `refine_content` function.

Changes include:
*   Adding explicit introductory text for the Markdown, individual images, and PDF sections to guide the LLM on the purpose of each input.
*   Introducing clear "START OF IMAGE" and "END OF IMAGE" delimiters for each image to better define their boundaries.
*   Unifying the PDF attachment mechanism for both Gemini and OpenAI providers, simplifying the code and ensuring consistent handling of PDF input.

These changes aim to improve the LLM's understanding of the provided content, leading to more accurate and relevant refinements.

2025-11-12 19:14:19 +11:00

8.5 KiB

Executable File

Raw Permalink Blame History

View Raw