seo-report
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes research data from other tools to generate reports, which represents a potential indirect injection surface. However, this is well-mitigated by strict output constraints.
- Ingestion points: Research data and recommendations passed from preceding skills.
- Boundary markers: Not explicitly defined for the input data stream.
- Capability inventory: Writing files to the project via the
save_reporttool. - Sanitization: The skill implements high-quality sanitization rules for the generated HTML, including a total ban on
<script>tags, a block on all external network resources (images, fonts, scripts), and a prohibition on JavaScript template literal syntax (backticks and${) to prevent injection into the platform's viewer environment.
Audit Metadata