Open Router
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches model metadata and performs inference requests using official OpenRouter API endpoints at openrouter.ai.
- [COMMAND_EXECUTION]: Provides utility commands using
curlandjqinauth-and-provider.mdandfallback-reliability.mdto verify network connectivity and authentication status. - [DATA_EXFILTRATION]: Transmits user prompts to OpenRouter for model execution, which is the primary intended purpose of the skill. It correctly recommends using environment variables for secrets rather than hardcoding them.
- [SAFE]: The skill maintains its own state and configuration files within a dedicated local directory
~/open-router/, avoiding interference with sensitive system files.
Audit Metadata