yolo

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

• Third-party content exposure detected (high risk: 0.90). The skill's browser automation explicitly navigates to the third‑party site lovable.dev and production URLs and uses read_page/find/DOM queries to read and parse user-generated content (GitHub commit messages, Lovable chat/log responses, and site consoles) as part of its workflow, which exposes the agent to untrusted third-party input that could carry indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly navigates at runtime to the user's Lovable project URL (e.g., https://lovable.dev/projects/abc123) and reads DOM/chat content (commit messages, assistant responses, secrets) to decide what prompts to submit and how to proceed, so this external URL is a required runtime dependency that directly controls agent behavior.