seo-audit
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, unauthorized data exfiltration, or obfuscated content were detected across the skill files.
- [DATA_EXPOSURE]: The skill attempts to access local configuration files like .agents/product-marketing-context.md to provide relevant SEO advice. This is a legitimate functional requirement for establishing context.
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves analyzing external website content, which acts as an ingestion point for untrusted data. Ingestion points: target website content fetched via web tools; Boundary markers: not present; Capability inventory: utilizes web_fetch and curl for data acquisition; Sanitization: no explicit sanitization described in the instructions. This represents a standard surface for indirect prompt injection inherent to the tool's purpose.
Audit Metadata