PDF Extraction

PDF workflows can now return complete document content alongside the fields and tables you extract.

• Read PDFs with selectable text, scanned pages, or a mix of both.
• Keep document content as HTML, Markdown, or both; including headings, tables, and captions.
• Include images as linked attachments, or omit them when you only need text and tables.
• Combine PDF data with details from the web page that linked to it, such as the title, publication date, and source URL.

Start with a direct link to one or more PDFs, or a web page with links to one or more PDFs. Tell the Kadoa Agent what to collect and what each row should represent, then review the preview before approving.

If these output options are unavailable in your workspace, contact the Support Center.

PDF extraction documentation →