html-ppt-zhangzara-creative-mode
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it instructs the agent to fill an HTML template with untrusted user data without specifying sanitization or boundary markers.
- Ingestion points: User-provided content such as headlines, body copy, names, and dates are used to replace placeholders in the
example.htmltemplate. - Boundary markers: Absent; there are no instructions to use delimiters or ignore embedded instructions within the user-provided data.
- Capability inventory: The skill produces an HTML artifact that includes a local JavaScript navigation script (
assets/deck-stage.js). - Sanitization: Absent; the instructions do not require the agent to validate or escape user input before inclusion in the final HTML.
- [EXTERNAL_DOWNLOADS]: The
example.htmltemplate fetches font assets from well-known Google services (fonts.googleapis.com and fonts.gstatic.com) to implement the design system's typography.
Audit Metadata