docx-to-md
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process external Word (.docx) documents, which creates a potential surface for indirect prompt injection where a document could contain hidden instructions to influence agent behavior.
- Ingestion points: The skill ingests .docx files provided by the user or an external source.
- Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore instructions embedded within the document content.
- Capability inventory: The skill executes a local Node.js script (.github/muscles/docx-to-md.cjs) and relies on the
pandocsystem utility for conversion. - Sanitization: No specific sanitization or validation logic for the input document content is described in the provided files.
- [NO_CODE]: The referenced logic script
.github/muscles/docx-to-md.cjswas not included in the provided file set, so its internal implementation could not be audited. Analysis is limited to the markdown instructions and JSON metadata.
Audit Metadata