heavy-file-ingestion-claude-code

To convert large PDF files to markdown, run the bundled deterministic converter script which produces compressed, model-friendly markdown artifacts and a generated index.md to guide further analysis. This prevents wasting model context on raw binaries.

The best way to extract tables from large spreadsheets is using a converter that exports per-sheet CSVs for XLSX, CSV, and TSV files. This creates compact, structured artifacts and an index.md that guide efficient model reads.

Claude Code wastes context on large DOCX files because it attempts to process raw binary data directly. Converting heavyweight files into compressed markdown artifacts first prevents this waste and enables index-driven analysis.

Yes, you need Python installed to run the deterministic converter script. The script prefers installed Python dependencies for converting PPTX and other formats, with fallbacks available, saving converted artifacts next to source files.

Yes, you can summarize large PDF content by converting it with the bundled script to generate markdown and an index.md. Read the index first to decide which extracted sections or tables to summarize, avoiding full file reads.