skills/cosmix/loom/loom-kubernetes/Gen Agent Trust Hub

loom-kubernetes

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard Kubernetes CLI operations via kubectl and helm. These commands are used for resource deployment, troubleshooting (e.g., kubectl logs, kubectl exec), and cluster inspection.
  • [EXTERNAL_DOWNLOADS]: Includes references to the Bitnami Helm repository (https://charts.bitnami.com/bitnami) for managing chart dependencies. Bitnami is a well-known and reputable source for Kubernetes packages.
  • [CREDENTIALS_UNSAFE]: Contains example Kubernetes Secret manifests with placeholder authentication strings, such as database-url: "postgresql://user:password@db-host:5432/myapp". These are provided as instructional templates, and the skill includes specific best practices advising users to never hardcode actual credentials.
  • [DATA_EXFILTRATION]: Provides commands to retrieve sensitive data from the cluster environment, such as kubectl get secret ... -o jsonpath='{.data}', for administrative and troubleshooting purposes. This is standard functionality for Kubernetes operations.
  • [SAFE]: The skill's instructions strongly emphasize security hardening, including running containers as non-root, dropping Linux capabilities, and implementing zero-trust network policies.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 07:34 AM
Security Audit — agent-trust-hub — loom-kubernetes