ml-ops-engineer

Warn

Audited by Snyk on Mar 13, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.80). The Kubernetes manifest references external container images (e.g., gcr.io/project/model-server:v1.2.3) which are pulled at runtime and execute remote code, and the skill also relies on external model/storage URIs (e.g., s3://models/production/v1.2.3, gs://feature-store/registry.db) as required dependencies—so these external URLs are runtime sources that execute code/artifacts.

Issues (1)

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 13, 2026, 01:20 AM
Issues
1