multi-search-engine
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: No shell commands or script execution patterns were detected. The skill primarily provides configuration and documentation for search URL templates.
- [EXTERNAL_DOWNLOADS]: The skill references well-known search engines and technology services (e.g., Google, DuckDuckGo, WolframAlpha, Brave). These are recognized as safe, well-known services.
- [DATA_EXFILTRATION]: No sensitive data access or exfiltration techniques were found. The documentation includes examples of searching for sensitive patterns (e.g., 'intext:password filetype:txt') as part of an advanced search guide, which is a common OSINT technique and aligned with the skill's function.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external web content (search results) into the agent's context. While this is an inherent attack surface for search-based skills, the skill itself provides only templates and does not implement exploitable processing logic.
Audit Metadata