docx

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime path reads outsider-authored free text by extracting/parsing the user-supplied DOCX contents (e.g., unpack.py/pandoc operate on document.docx), which can include arbitrary strings embedded in word/document.xml that the workflow then ingests for analysis/editing.