frame-liquid-bg-hero
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references Tailwind CSS from
cdn.tailwindcss.comand typography fromfonts.googleapis.com. These are well-known services used to provide the styling and font assets for the hero frame. - [INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text (quotes and titles) to generate a visual frame, which could potentially render malicious or misleading instructions if the input is not treated as untrusted data.
- Ingestion points: The instructions in
SKILL.mddirect the agent to use user-provided text to populate the primary header and subtitle fields in the HTML output. - Boundary markers: The skill lacks instructions for using delimiters or explicit "ignore embedded instructions" warnings when processing user input.
- Capability inventory: The generated output is a static visual HTML/CSS/Canvas frame; the implementation does not utilize sensitive capabilities such as file system writes or unauthorized network requests.
- Sanitization: There are no explicit instructions to escape or sanitize the user-provided text before it is interpolated into the HTML template.
Audit Metadata