try

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

• Third-party content exposure detected (low risk: 0.10). The workflow instructs the agent to pin targets, fetch READMEs/quickstarts, clone repositories, and run code or skills from arbitrary user-supplied target names, package names, or repo/docs URLs.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to clone, install, and execute arbitrary libraries, frameworks, CLI tools, and repositories specified dynamically by the user, representing a weak/unknown/suspicious runtime dependency pattern.