market-emails

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and analyze a provided URL ("When This Skill Is Invoked" / Phase 1: "If a URL is provided, fetch the site to understand the business, product, audience, and voice"), which clearly ingests untrusted public website content that the agent is expected to read and use to drive email-writing decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly fetches the user-provided site at runtime (the URL passed to /market emails <topic/url>) and uses the fetched site copy to inform and directly control the generated email prompts and content, so the external URL content can alter agent instructions.