explainer-visuals
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [External Downloads] (SAFE): The skill references industry-standard libraries (D3.js, GSAP) via CDNs.\n
- Mitigation: It explicitly instructs the agent to include integrity hashes, which prevents execution of tampered remote code.\n- [Indirect Prompt Injection] (LOW): The skill processes untrusted user data to generate executable code.\n
- 1. Ingestion points: User-provided essay content and conceptual descriptions enter the agent context through the skill's triggers.\n
- 2. Boundary markers: Absent; the skill does not define clear delimiters to separate user data from the generated script logic.\n
- 3. Capability inventory: The skill outputs self-contained HTML/JavaScript files with browser-level execution capabilities.\n
- 4. Sanitization: Absent; no instructions are provided to sanitize or escape user input before it is embedded into the output code, posing a potential XSS risk.
Audit Metadata