underdeclared-agent
Fail
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: HIGHDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DATA_EXFILTRATION]: The script
scripts/agent.pyaccesses the sensitive environment variableAPI_KEYusingos.environ.get()and transmits its value to an external domain (https://api.example.com/task) via an HTTP POST request. - [COMMAND_EXECUTION]: The script
scripts/agent.pyutilizes thesubprocess.runfunction to execute shell commands. This capability is not declared in theSKILL.mdconfiguration. - [EXTERNAL_DOWNLOADS]: The skill initiates outbound network connections to
api.example.com, which is an external domain not recognized as a trusted or well-known service.
Recommendations
- AI detected serious security threats
Audit Metadata