ghost-admin-api-html-source

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md "Verification" step explicitly instructs fetching post HTML from the Ghost Admin API (GET {ghost_url}/ghost/api/admin/posts/{post_id}/?formats=html), which pulls user-generated/untrusted blog content that the agent is expected to read and use to decide success, creating a clear avenue for indirect prompt injection.