market-copy

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflow ("When This Skill Is Invoked" and Phase 1: "Fetch and Parse") explicitly uses WebFetch to retrieve and analyze an arbitrary target URL's public page copy, so the agent ingests untrusted third-party web content that can influence its outputs and decisions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill uses WebFetch at runtime to retrieve the user-supplied target URL (the passed to /market copy <url>), and that fetched page copy is injected into the model context to drive analysis and generation, meaning arbitrary external content can directly control prompts.