observability-k8s-investigation
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Technical analysis shows the skill is restricted to observability tasks using standard OpenTelemetry and Elastic patterns. The provided ES|QL recipes and API references are consistent with legitimate diagnostic use cases.
- [INDIRECT_PROMPT_INJECTION]: The skill processes application logs and Kubernetes events, which are external data sources. This constitutes an expected surface for a diagnostic tool. Ingestion points include
logs-*.otel-*indices. No boundary markers or sanitization logic are defined in the provided queries, but agent capabilities are limited to read-only analysis and report synthesis, presenting no significant risk. No other security concerns were identified.
Audit Metadata