word
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements professional document processing workflows. It uses security-focused libraries like
defusedxmlto handle XML parsing, which prevents XML External Entity (XXE) and expansion attacks. - [COMMAND_EXECUTION]: The skill uses
subprocess.runto interface withsoffice(LibreOffice),pdftoppm, andgit. These calls are essential for rendering documents to images for visual review and performing word-level diffs for tracked changes. All subprocess calls use list-based arguments, mitigating the risk of command injection through manipulated filenames. - [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted document content (an ingestion surface), it does so using structural editing tools. The focus is on OOXML manipulation and visual verification rather than autonomous decision-making based on document text, and the use of
defusedxmlprovides a safe layer for processing the XML structure.
Audit Metadata