Use the groupdocs-parser command to extract text, document info, metadata, Images and annotations from documents in a shell, a Make rule or a CI step, without writing Python code....you can pull text, metadata, images and annotations out of documents...value images FILE --out DIR Saves the embedded images as image-001...