This project currently inputs the first page page of a PDF, runs the text through the GPT-3.5-Turbo, and then returns a PDF with metadata tagged in the 'Keywords' field.
- Save a PDF file to your computer
- When running main.py, the program will prompt you for the file path of the PDF you would like to tag (TODO Edit this section)
- It will then prompt you for an output file path for the GPT model to run though
- When you open the output file, it will have metadata tagged in the 'Keywords' field
- Put the output keywords in obsidian
- Get citation information from input pdfs
- Rip metadata/citation information form the pdf
- Create Markdown file with same file name as PDF (for example: bunny.pdf -> bunny.md)
- Paste citation information + keywords as tag + metadata in markdown for each file
EXAMPLE:
filename: bunny.md