heavy-file-ingestion

To convert PDFs, DOCX, PPTX, and spreadsheets into markdown, you can run a Python converter that transforms heavyweight files into compact artifacts before model reasoning occurs. This process yields markdown or per-sheet CSVs plus a lightweight index.

Normalizing an XLSX workbook involves converting multi-sheet files into per-sheet CSVs and generating an index manifest. This approach surfaces structure and quality flags so the main agent can decide whether to read the extracted artifacts or escalate.

Yes, the file conversion process works with PDFs, DOCX, and PPTX files, utilizing optional libraries such as pdfplumber, python-docx, and python-pptx. It extracts text, tables, and slide outlines into markdown artifacts.

Yes, converting a large CSV or TSV generates an index.md and index.json containing previews, stats, and quality warnings. These outputs include a recommended next step to help your workflow decide whether to process the artifact or escalate.

You need the bundled Python converter for basic extraction, but optional libraries like pdfplumber, python-docx, python-pptx, and openpyxl are recommended for handling complex PDFs, DOCX, PPTX, and XLSX files effectively.

You should avoid reading raw PDFs and use file conversion when agents risk wasting model context and cost on large structured files. Converting them first into compact markdown prevents inefficient context usage during deep analysis.