seo
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill orchestrates complex SEO audits by executing internal Python scripts via a custom 'claude-seo run' launcher, ensuring environment isolation.
- [EXTERNAL_DOWNLOADS]: Fetches SEO metrics and mapping data from well-known providers like Google, Moz, Bing, and OpenStreetMap.
- [PROMPT_INJECTION]: Vulnerable to indirect prompt injection as it processes raw content from external URLs. Ingestion points: URLs processed by audit and analysis commands. Boundary markers: No explicit delimiters or instructions are provided to the agent for content isolation. Capability inventory: Access to network utilities (curl) and internal script execution. Sanitization: Not explicitly documented.
- [CREDENTIALS_UNSAFE]: References checking for API keys (e.g., Google, Moz) through specific scripts, prioritizing environment-based credential management over hardcoding.
Audit Metadata