docx-to-web-publisher
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a legitimate workflow for converting document content into styled React components. It relies on standard system utilities (ImageMagick, Pandoc) and a Python script using the standard library. No evidence of malicious code, data exfiltration, or unauthorized persistence was found.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted .docx files, which represents an ingestion surface for indirect prompt instructions.
- Ingestion points: The
scripts/extract_docx.pyscript and the extraction workflow inSKILL.mdingest text from user-provided.docxfiles. - Boundary markers: No explicit delimiters or instructions to ignore embedded malicious text are provided in the extraction process.
- Capability inventory: The skill instructs the agent to create new files in the
src/content/directory, update project configuration files, and execute local shell commands (python3,magick,pandoc). - Sanitization: The extracted text is mapped to JSX components without explicit sanitization, though the manual or templated nature of the conversion provides a natural layer of review.
Audit Metadata