exploiting-containers

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The skill document outlines container security assessment and escape methodologies, including downloading enumeration scripts from GitHub (e.g., deepce), which constitutes low-risk ingestion of external source content.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading an executable shell script from an unverified GitHub repository (https://github.com/stealthcopter/deepce/raw/main/deepce.sh) and executing it during runtime, presenting a high-risk untrusted operational dependency.