seo-fundamentals
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No instructions designed to bypass safety or override agent behavior were found.
- Data Exposure & Exfiltration (SAFE): The skill does not access sensitive system paths or credentials, and contains no network-capable code.
- Obfuscation (SAFE): No hidden or encoded malicious patterns detected.
- Unverifiable Dependencies & RCE (SAFE): The script uses standard libraries and does not download or execute remote code.
- Indirect Prompt Injection (LOW): The script 'scripts/seo_checker.py' reads untrusted data from HTML and JSX files. Evidence: 1. Ingestion: find_pages() reads project files. 2. Boundaries: Absent. 3. Capabilities: Read-only static analysis. 4. Sanitization: Absent. No dangerous capabilities are exposed to this data.
- Dynamic Execution (SAFE): No dynamic code generation or execution functions are used.
Audit Metadata