document-reader

To ingest local PDFs and images for agent processing, use this Skill to read files in binary mode, enabling OCR and vision tasks to extract text from scanned documents or structured data.

Yes, this document reader supports parsing structured data formats like JSON and CSV. It offers text mode for data extraction with configurable options for structured data parsing and size limits.

Yes, you can use binary mode for OCR on scanned documents. This mode captures image data necessary for vision-based processing, allowing agents to read and extract text from scanned files.

The best way to ingest local files without custom file I/O code is using this Skill, which abstracts file input operations and standardizes ingestion for PDFs, images, JSON, CSV, and plain text into agent workflows.

When processing large local files, limitations apply through configurable size limits for parsing and ingestion. You must set appropriate size boundaries to manage reading operations for large PDFs, images, or structured data formats.