create-an-asset
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data and using it to generate rendered HTML output.
- Ingestion points: User-uploaded materials (transcripts, emails, notes, call recordings) and automated web research results (Phase 1).
- Boundary markers: The skill lacks explicit instructions to treat external data as untrusted or to use delimiters to prevent the agent from executing instructions hidden in research content.
- Capability inventory: The skill is designed to generate self-contained HTML files that include inline JavaScript and CSS, which could be exploited to include malicious scripts if those scripts were injected into processed data.
- Sanitization: There is no mention of sanitizing or escaping the content extracted from external sources (such as the "Key quotes" mentioned in Phase 3) before interpolating it into the final asset, creating a potential Cross-Site Scripting (XSS) risk for the consumer of the asset.
Audit Metadata