The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/agent.py executes AWS CLI commands using subprocess.run() with a list of arguments. This approach is secure as it prevents shell injection by avoiding shell interpolation, allowing the agent to interact with the GuardDuty service safely.\n- [SAFE]: The skill's ingestion of GuardDuty findings is handled through structured data parsing. The scripts/agent.py script extracts specific metadata (e.g., finding type, severity, resource ID) and truncates long descriptions, minimizing the risk from potentially malicious data in the findings.\n- [SAFE]: No evidence of hardcoded credentials, suspicious external downloads, or attempts to establish persistence was found. The workflow and tools provided are consistent with the stated goal of cloud security operationalization.

detecting-cloud-threats-with-guardduty