on-page-seo-auditor
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted data from external URLs using the WebFetch tool. It includes explicit boundary instructions directing the agent to treat all fetched content as passive data and to ignore any directives, meta-tags, or HTML comments that attempt to influence the audit's behavior or override system instructions.
- [DATA_EXFILTRATION]: The skill maintains audit logs and results within a dedicated memory/ directory structure. It does not attempt to access sensitive system files (e.g., credentials or configuration files) or exfiltrate information to third-party servers beyond the requested SEO analysis scope.
- [SAFE]: No obfuscation, unauthorized command execution, or remote code patterns were detected. All external resource references are consistent with the author's verified GitHub organization.
Audit Metadata