ads-landing

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly requires using WebFetch to fetch the provided URL and "Extract all visible text content" from arbitrary landing pages (Execution Flow step 1 and Rules 1), so the agent ingests and acts on untrusted third-party web content which could contain injected instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly uses WebFetch at runtime to retrieve the user-provided landing page URL (the external URL passed to /ads landing ), and the fetched page content is injected into the audit/rewrite process—allowing remote page content to directly influence and control the agent's prompts and outputs.