seo-specialist

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists exclusively of markdown instructions and static JSON assets. There are no executable scripts (.sh, .py, .js) or binaries included, which significantly limits the potential for malicious behavior.
  • [PROMPT_INJECTION]: The instructions do not contain any patterns typical of prompt injection, such as 'ignore previous instructions' or attempts to bypass safety filters. On the contrary, the skill explicitly mandates 'white-hat' and 'ethical' SEO practices.
  • [DATA_EXFILTRATION]: No sensitive file paths, hardcoded credentials, or network exfiltration patterns were found. The 'allowed-tools' configuration is empty, which restricts the agent's ability to use potentially dangerous tools.
  • [EXTERNAL_DOWNLOADS]: There are no commands or configurations that attempt to download code or data from external URLs or package registries.
  • [COMMAND_EXECUTION]: No shell commands or dynamic execution patterns were identified in the instructional body or the JSON assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 04:15 AM
Security Audit — agent-trust-hub — seo-specialist