ralph-deep-planner

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

• Third-party content exposure detected (high risk: 0.90). The workflow's "agent-browser 시각 검증" step (Phase 3/Phase별 작업 and the "agent-browser 시각 검증" section) explicitly runs agent-browser open and evaluates page content (snapshots, console errors), meaning the agent will fetch and interpret arbitrary web pages (potentially untrusted third‑party content) and use those observations to drive fixes and next actions.