cluster-report
Warn
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill contains 'CRITICAL Script Rules' that explicitly instruct the AI agent to never read the source code of its helper scripts,
assemble.pyandaggregate.py. This is a direct attempt to bypass the agent's safety reasoning and prevent it from performing security analysis on the code it is required to execute. - [COMMAND_EXECUTION]: The workflow requires the execution of shell commands (
mkdir) and Python scripts (python3 ocp-admin/scripts/cluster-report/assemble.py) on the host system to aggregate cluster data. - [DATA_EXFILTRATION]: The skill is designed to collect and aggregate sensitive environment metadata, including node resources, namespace lists, and pod statuses, from multiple OpenShift or Kubernetes clusters into temporary manifest files. While it warns against exposing kubeconfig credentials, the consolidation of fleet-wide metrics represents a data exposure surface.
Audit Metadata