ae-use-agent

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow instructs the agent to fetch and execute instructions, skill contents, and scripts from an AE agent platform via the ae-cli tool (+get-agent-context, +get-skill-content), where agents and skills can be authored by arbitrary platform users.

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

• Potentially malicious external URL detected (high risk: 0.70). The npm registry URL in references/local-agent.md is an explicit package-manager registry locator pointing to a private internal domain, which constitutes an unverifiable operational dependency under the dependency gate.