ln-92-container-engineer
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external project files such as container manifests and lockfiles to perform build operations.
- Ingestion points: Container manifests, CI configurations, and project lockfiles identified in SKILL.md.
- Boundary markers: Mandates citing blocking instructions and reconciling tasks against observed evidence before completion.
- Capability inventory: Authorized for container image builds, local runtime packaging, and isolated startup/health verification.
- Sanitization: Explicitly requires verifying the absence of secrets in image metadata, layers, and logs, and enforcing least-privileged runtime identities.
Audit Metadata