product-marketing-context
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the agent-browser tool from the official Vercel Labs GitHub repository. It also suggests installing the agent-browser package via npm. These sources are recognized as trusted, and the installation steps are standard for providing necessary browser automation capabilities for research tasks.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted external data from live website URLs to perform marketing audits.\n
- Ingestion points: Public URLs visited via agent-browser, WebFetch, or WebSearch as referenced in shared-patterns.md.\n
- Boundary markers: Absent; there are no instructions to delimit external content or warn the agent about potential embedded instructions.\n
- Capability inventory: The skill can write files to the local brand directory and perform further network research based on analysis.\n
- Sanitization: Absent; the skill does not perform validation or escaping of the ingested web content before using it for document generation.
Audit Metadata