md-to-modern-pptx
Pass
Audited by Gen Agent Trust Hub on Feb 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary logic in both JavaScript and Python scripts is purely functional, handling file processing and API communication in accordance with standard development practices.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it uses content from user-provided Markdown files to build prompts for image generation without sanitization. This is a common architectural pattern for such tools and is not inherently malicious.
- Ingestion points: User-supplied Markdown files read by
scripts/gemini_image_pack.py. - Boundary markers: Content is directly interpolated into the prompt string in
prompt_templatewithout delimitation. - Capability inventory: Authenticated network communication with AI gateways using API keys provided by the user.
- Sanitization: No escaping or filtering is applied to the Markdown text prior to its use in generating image prompts.
Audit Metadata