dmux-workflows

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md "Pattern 1: Research + Implement" explicitly instructs a pane to "Research best practices... Check current libraries" (i.e., ingest public web/library information) and then merge those findings into other panes, meaning untrusted third‑party content could be read and materially influence subsequent agent actions.