doc-reader

To extract text from a scanned PDF or image, you need optical character recognition to identify characters. This skill uses rapidocr-onnxruntime to perform OCR, ensuring high-fidelity data retrieval from image files and scanned documents.

The best way to extract structured data from Office files and PDFs is using a unified parsing interface that handles diverse formats. This skill processes Word, Excel, PowerPoint, and PDF documents, returning a consistent JSON envelope regardless of the source format.

Yes, you can parse a spreadsheet without leaving your chat environment. This skill extracts data and text from Excel files directly within your interface, previewing contents and simplifying downstream processing without manual file conversion.

Yes, this text extraction tool supports source code and config files. It implements universal parsing to extract text from various code files, alongside administrative documents like contracts, using robust encoding fallbacks to ensure accurate data retrieval.

Document parsing returns a JSON envelope format to provide a unified output structure. This consistent JSON format simplifies downstream processing, allowing you to easily summarize contracts or analyze spreadsheet data extracted from heterogeneous file types.