ralph-loop

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

• Third-party content exposure detected (high risk: 0.70). The skill includes curl commands that fetch markdown from public URLs on fullstackrecipes.com (e.g., https://fullstackrecipes.com/api/recipes/agent-setup, user-stories-setup, ralph-setup), so the agent would ingest content from an external public website that could be untrusted or user-controlled.