duplicate-description
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is instructional in nature and provides best practices for SEO auditing. It does not include executable scripts, shell commands, or requests for sensitive system permissions.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill references official documentation from Google Search Central and MDN Web Docs, as well as the community resource Frontend Checklist. It suggests using standard SEO tools like Screaming Frog and Sitebulb. All external references are to well-known, legitimate services and do not represent a data exfiltration risk.
- [INDIRECT_PROMPT_INJECTION]: The skill involves auditing external HTML content. While this represents a potential ingestion surface for untrusted data, the skill's capabilities are limited to providing analysis and advice. 1. Ingestion points: External website HTML scanned for meta tags (SKILL.md). 2. Boundary markers: None specified. 3. Capability inventory: No file-write, subprocess execution, or network exfiltration capabilities are present in the skill's own code or instructions. 4. Sanitization: Not applicable as no automated processing logic is provided. The risk is inherent to the task of web auditing and is addressed by the agent's standard safety guidelines.
Audit Metadata