skills/samaeldemiurgos2-cyber/google-skills/gke-ai-troubleshooting-jobset-interruption/Gen Agent Trust Hub
gke-ai-troubleshooting-jobset-interruption
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes Kubernetes container logs to troubleshoot workload failures.
- Ingestion points: Worker container logs are read via Cloud Logging filters in Step 4 and evaluated against signatures in
references/failure_signatures.md. - Boundary markers: The prompt templates do not include specific delimiters or instructions to ignore potential commands embedded within the logs.
- Capability inventory: The skill environment allows shell command execution (via
scripts/validate_queries.sh) and provides instructions for the agent to recommend or trigger the deletion of GCE VM instances. - Sanitization: The skill lacks explicit sanitization or filtering logic for the raw log content before it is processed by the agent.
- [COMMAND_EXECUTION]: The
scripts/validate_queries.shfile contains executable bash code that invokesgcloudandcurlto interact with cloud environment resources. - [CREDENTIALS_UNSAFE]: The validation script programmatically acquires sensitive identity tokens using
gcloud auth print-access-tokento authenticate requests to external monitoring APIs.
Audit Metadata