docx-to-web-publisher

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill describes a workflow to process user-supplied .docx files and convert them into React components, which creates a surface for indirect prompt injection where malicious instructions inside the document could influence the agent's code generation.\n
  • Ingestion points: Untrusted data enters the agent context via the scripts/extract_docx.py script and the extraction commands in SKILL.md.\n
  • Boundary markers: The instructions do not define explicit delimiters or instructions to the agent to disregard potential instructions embedded in the document text.\n
  • Capability inventory: The skill workflow includes running shell commands for image conversion (magick) and site builds (npm run build).\n
  • Sanitization: No sanitization or validation logic is provided to filter malicious content from the extracted text before it is used to generate React components.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 01:28 PM
Security Audit — agent-trust-hub — docx-to-web-publisher