setup-k8s-collection

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill generates shell commands for Kubernetes and Helm operations. It includes explicit instructions ensuring these commands are presented to the user for local execution, which prevents the agent from attempting to perform privileged operations within its own sandbox.
  • [EXTERNAL_DOWNLOADS]: The skill references Helm chart repositories at https://observeinc.github.io/helm-charts and official OpenTelemetry resources. These are verified vendor-managed sources required for the deployment process and do not involve untrusted third-party code.
  • [CREDENTIALS_UNSAFE]: The skill requests an ingest token from the user for authentication. It provides a secure handling method by generating a kubectl create secret command with a placeholder, ensuring the sensitive token is stored correctly as a Kubernetes secret in the user's infrastructure without being processed or stored by the agent itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:46 PM
Security Audit — agent-trust-hub — setup-k8s-collection