The specific problem with PDF files
A PDF stores glyph positions, not sentences. Extraction has to reconstruct reading order, and that is where it goes wrong: two-column academic layouts interleave, headers and footers repeat on every page, and table rows lose their alignment with the column headers above them.
- Open the extracted text and confirm paragraphs read in the right order, especially in multi-column sections.
- Strip the running header and footer, which otherwise repeat once per page and waste real tokens.
- Check that tables kept their row and column relationship, or replace them with a written summary.
- Confirm there is a text layer at all. If the extraction is empty, the file is a scan and needs OCR first.
What LLM-friendly text means
LLM-friendly text is not just extracted text. It is labeled, grouped, and ordered so the model can tell where each section came from and how it relates to the task.
Riflet focuses on building that context file from mixed sources, not only converting one isolated file.
Workflow
Use the conversion as part of a complete context bundle.
- Add your PDF files to Riflet.
- Add supporting notes, briefs, or source files if they explain the document.
- Remove sections that do not matter for the question.
- Export Markdown or text for your AI tool.
Quality checks
Text-based PDFs work best. Scanned PDFs need OCR before any AI tool can reason over their contents reliably.
After export, skim the headings and source boundaries. The best bundle should make sense to a human before it goes to an AI.
Build your AI context file with Riflet
Select the sources that matter, check the token budget, and export one clean file for the AI you already use.
Download Riflet freeFrequently asked questions
Can I use this PDF conversion bundle in more than one AI tool?
Yes. Riflet exports plain text or Markdown, so the same context file can be used in Claude, ChatGPT, Gemini, Perplexity, or a local model if that tool accepts pasted text or file uploads.
Does Riflet automatically update the AI after my files change?
No. Riflet is deliberately export-based. When your files change, reopen the workspace, let Riflet reapply the saved sources and filters, export a fresh bundle, and upload it where you want to use it.
How do I avoid giving AI too much context?
Start with the decision or task, then include only the files that would help a smart human answer it. Use Riflet's token count to remove stale, duplicated, generated, or background-only files before exporting.
Do my files leave my computer when Riflet builds the bundle?
Riflet processes selected files locally. Your content leaves your machine only if you choose to paste or upload the exported bundle into an AI product.