anthropic-pdf

To extract text and tables from PDF documents, this Skill uses Python libraries like pdfplumber and pypdf to parse structured content, metadata, and scanned-page content. It reliably processes invoices by pulling tabular data and text for downstream automation workflows.

Yes, you can fill PDF forms and add text annotations. The Skill detects fillable form fields, validates bounding boxes, and applies text annotations for non-fillable documents, enabling automated form completion without manual copy-paste.

You need pypdf, pdfplumber, reportlab, pdf2image, and pillow to merge, split, and generate PDF files. These dependencies support rendering, cropping, watermarking, encrypting, and rotating PDF documents within automation scripts.

The best way to automate PDF form completion is using field detection scripts with bounding box validation. This Skill handles fillable forms directly and overlays text annotations on non-fillable documents, ensuring accurate field population.

Yes, this approach works for extracting metadata and rendering scanned PDF pages. It uses pdf2image and pillow to convert pages into images, allowing you to extract content from scanned documents and process metadata alongside structured text.

PDF table extraction fails on scanned documents because scanned pages contain images rather than selectable text. This Skill uses pdf2image and pillow to render pages into images first, enabling reliable content extraction from scanned PDFs.