kubectl-debugging

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill extensively documents the use of kubectl debug to initiate interactive troubleshooting sessions. These commands allow for executing shells and diagnostic tools within containerized environments and on host nodes, which is the primary administrative purpose of the skill.
  • [EXTERNAL_DOWNLOADS]: References several common and well-known container images for debugging, including busybox, alpine, ubuntu, and nicolaka/netshoot. These are standard industry tools for DevOps and cluster administration.
  • [DATA_EXFILTRATION]: Provides instructions for node-level debugging that grants access to the host filesystem (at /host) and system logs. While this involves access to sensitive host data, it is a documented and intended feature of Kubernetes for cluster-level diagnostics.
  • [PROMPT_INJECTION]: The skill uses tools like WebFetch and stern to ingest external data such as web content and container logs. This represents a potential surface for indirect prompt injection where malicious instructions could be embedded in the data being debugged.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 09:33 PM
Security Audit — agent-trust-hub — kubectl-debugging