agenty
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes the @membranehq/cli Node.js package for setup and authentication. This is an official vendor resource from membranedev and is considered a safe dependency for the intended integration logic.
- [COMMAND_EXECUTION]: The skill instructions include several calls to the membrane CLI, such as membrane login, membrane connect, and membrane action run. These are used to manage the lifecycle of the Agenty integration and do not involve arbitrary command execution or privilege escalation.
- [SAFE]: No prompt injection, data exfiltration, or obfuscation patterns were detected. The skill correctly leverages the vendor's proxy system to handle Agenty credentials, ensuring that API keys are not exposed or managed insecurely by the user.
Audit Metadata