docx-render
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on executing shell commands to invoke Python scripts, manage Docker containers, and establish SSH connections to a rendering guest.\n
- Evidence: Use of
python3 scripts/doc_render.py IN.docx OUT.pdf --renderer word-remotefor document conversion.\n - Evidence: Commands like
docker start omarchy-windowsandssh -o ConnectTimeout=5 word@winvm exitare used to manage the remote rendering environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external document files (.docx,.doc,.xlsx,.pptx) which could contain malicious content or metadata designed to influence the agent's behavior during the rendering process.\n - Ingestion points: User-provided input files (
IN.docx,IN.doc,IN.xlsx,IN.pptx) are passed to scripts inSKILL.md.\n - Boundary markers: No explicit boundary markers or "ignore embedded instructions" warnings are documented for the rendering process.\n
- Capability inventory: The skill can execute shell scripts, interact with Docker, and perform SSH operations as documented in
SKILL.md.\n - Sanitization: The documentation does not specify how input filenames or document contents are sanitized before being passed to the underlying rendering engines.
Audit Metadata