WiseDiag Medical Report Interpretation

Problem It Solves

Medical reports often contain test items, units, reference ranges, and abnormality flags. Users may need item-by-item explanations: what an elevated or low value means, which results deserve attention, and whether to consult a clinician. Passing report images directly to a model can still suffer from unstable OCR, missed key values, and difficulty comparing multiple reports. WiseDiag Report targets this workflow by sending local report images or PDFs to WiseDiag cloud services for recognition, parsing, and interpretation, then generating a Markdown result file.

How It Works and Boundaries

• Input modes are mutually exclusive: use -f or --file for local file paths, and --image for public image URLs; do not mix them in one command.
• Core capabilities: extract report text, parse key indicators, highlight abnormal values, and answer a custom prompt via --question, such as “explain abnormal items and provide advice.”
• Result output: results are saved by default in the current working directory as {name}.md; use -o to set the output directory and -n to set the filename.
• Runtime dependency: set the WISEDIAG_API_KEY environment variable. Requests go to an external API and may take longer for large files, so wait for the streaming output to complete.

Note: the skill sends report content to WiseDiag cloud services for processing, so avoid materials containing sensitive or private information that cannot be uploaded remotely. It supports JPG, JPEG, and PNG images as well as PDFs. Output is for reference only and does not constitute a diagnosis or treatment advice; health decisions should be made with a qualified healthcare professional.