create-lab-guide

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly clones and reads a public GitHub repo (https://github.com/dockersamples/{REPO_NAME}) and ingests files such as README.md, labspace/labspace.yaml, labspace/.md, .github/workflows/.yml, and compose.override.yaml — untrusted third-party content that the agent parses and uses to decide guide content and actions (e.g., model-download), enabling indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). Cloning https://github.com/dockersamples/{REPO_NAME}.git at runtime fetches repository content that is parsed and injected into the guide-generation process, meaning external repo files can directly control the agent's prompts/instructions.