container-security

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external sources such as Dockerfiles, Kubernetes manifests, and Helm charts.
  • Ingestion points: The skill utilizes Read, Grep, and Glob tools to locate and process files matching broad patterns (e.g., **/Dockerfile, **/*.yaml, **/*-rbac.yaml).
  • Boundary markers: A dedicated "Prompt Injection Safety Notice" is included in SKILL.md, explicitly instructing the agent to treat all string values, labels, and annotations as data and to disregard any directives embedded within those files.
  • Capability inventory: The skill's execution environment is strictly limited to read-only file access via allowed-tools. It lacks network access, file-write capabilities, or tools for arbitrary command execution.
  • Sanitization: The instructions provide a framework for evaluating configurations against static security benchmarks (CIS/NIST) rather than executing content found within the analyzed files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 01:49 PM
Security Audit — agent-trust-hub — container-security