pricing-advisor
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a vulnerability to indirect prompt injection that can lead to Cross-Site Scripting (XSS) in the results display.
- Ingestion points: Data entered by the user to describe their product is processed by the agent and passed to assets/webview.html through URL parameters.
- Boundary markers: There are no explicit instructions or markers to prevent the agent from including HTML or script tags in the generated pricing data.
- Capability inventory: The skill relies on a webview component to render the pricing strategy UI.
- Sanitization: While some fields in assets/webview.html use textContent, the tier details (name, target audience, and price) are inserted using innerHTML. This allows an attacker to craft a product description that tricks the agent into generating a script payload, which would then be executed in the webview context.
Audit Metadata