k8s-grafana-troubleshooting
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
kubectlto inspect the state of the Kubernetes cluster, including retrieving pod statuses, container logs, and resource configurations. It includes a specific command to perform a rollout restart of the Grafana deployment, which is correctly gated by instructions requiring explicit user approval before execution. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and interprets data from the Kubernetes cluster environment, such as logs and ConfigMaps, which constitutes an attack surface for indirect prompt injection.
- Ingestion points: Diagnostic data enters the agent context through
kubectl logs,kubectl get configmap, andkubectl eventsacross various cluster namespaces. - Boundary markers: No explicit delimiters or instruction-bypass warnings are used in the commands to wrap the data being read.
- Capability inventory: The skill is capable of executing deployment restarts and interactive queries via the Kubernetes API server.
- Sanitization: The skill does not sanitize the logs or event data, though it contains specific logic to filter out sensitive environment variables like passwords when querying resource specifications.
Audit Metadata