skills/jpcaparas/skills/seo-analysis/Gen Agent Trust Hub

seo-analysis

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions provide legitimate SEO auditing guidelines for codebase analysis without including malicious prompts or bypass techniques.
  • [COMMAND_EXECUTION]: The provided Python scripts in the scripts/ directory are intended for local development tasks such as regression testing and prompt building. They utilize standard library modules and do not include dangerous system calls or remote code downloads.
  • [DATA_EXFILTRATION]: No exfiltration patterns were detected. All network references in the documentation and metadata are to official technical documentation from established organizations like Google and the Open Graph protocol.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted codebase data during the audit workflow. It mitigates potential injection risks by instructing the agent to verify findings against rendered output and to prioritize evidence-backed claims over speculative instructions found in the source.
  • [OBFUSCATION]: The skill contains no encoded strings, hidden characters, or homoglyph-based attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 10:51 AM
Security Audit — agent-trust-hub — seo-analysis