doc-to-docx
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external
.docfiles provided by the user, which presents a surface for indirect prompt injection attacks where malicious instructions hidden in a document could influence agent behavior. - Ingestion points: Local file ingestion occurs in
scripts/doc2docx.pyvia theDocReaderclass, which reads the binary content of user-supplied.docfiles. - Boundary markers: Absent. The skill extracts and returns document text to the agent context without using specific delimiters or warnings to distinguish untrusted content from instructions.
- Capability inventory: The skill includes file system read access (via
olefile) and write access (viapython-docxandlxml) inscripts/doc2docx.py. It does not possess network access, arbitrary command execution, or privilege escalation capabilities. - Sanitization: Absent. The engine performs structural and formatting parsing but does not sanitize the extracted text for potential prompt injection patterns before passing it to the agent.
Audit Metadata