Cursor

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly allows and instructs Background Agents/Bugbot and remote workflows to clone/fetch repository code and PR metadata from https://github.com/* (see "Background Agents" in background-agents-and-bugbot.md and the External Endpoints table in SKILL.md), so the agent will ingest untrusted, user-generated third-party content that can materially influence execution and decisions.