infrastructure
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill integrates the official Grafana Helm repository (grafana.github.io/helm-charts) to fetch the k8s-monitoring chart. This is a well-known service for infrastructure monitoring components.
- [COMMAND_EXECUTION]: The instructions involve the use of standard DevOps tools such as kubectl, helm, and systemctl to manage monitoring agents and secrets. These actions are consistent with the primary purpose of cluster and host monitoring.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting untrusted telemetry data from external systems.
- Ingestion points: Monitoring data ingested through Kubernetes logs, container discovery, and cloud provider monitoring APIs (references/clouds-and-hosts.md).
- Boundary markers: The configurations do not explicitly include delimiters or instructions to ignore embedded commands in the telemetry data.
- Capability inventory: The skill has access to privileged commands such as helm install and kubectl create (SKILL.md).
- Sanitization: No sanitization logic for external logs or metric labels is defined in the provided templates.
Audit Metadata