google-search-ai-optimization
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The test suite in
scripts/test_skill.pyusessubprocess.runto execute the skill's internal validation and auditing scripts. These commands are constructed using static file paths for regression testing and do not involve unsanitized user input in a shell context. - [EXTERNAL_DOWNLOADS]: The
scripts/audit_page.pyutility utilizes the standardurlliblibrary to fetch content from URLs provided as command-line arguments. This behavior is essential for its core purpose of auditing live web pages for SEO compliance. The skill also references official documentation from well-known Google and Web.dev domains. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it processes external, untrusted HTML content via
scripts/audit_page.py. The tool uses a structural HTML parser to extract specific metadata and content signals, presenting this structured data to the agent for evaluation. No bypass or autonomous execution patterns were observed. - [SAFE]: The skill exhibits high-quality engineering practices, including a dedicated validation script and a clear separation between instructional guidance and functional tools. All file and network operations are restricted to the domain of web auditing.
Audit Metadata