agent-sources

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

• Third-party content exposure detected (low risk: 0.10). The required workflow instructs the agent to clone and read upstream package source code and tests from third-party repositories, which are authored by outsiders.

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 repositories from arbitrary user- or workspace-determined URLs during execution, constituting an unverifiable runtime external dependency.