vibe-task

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and semantically analyzes PR bodies, comments, and commit messages via commands like "vibe flow review --json" (see Step 1.2 and "Step 2.5: PR 语义分析"), using that untrusted, user-generated PR content to infer tasks and drive follow-up actions (task creation/association), which can materially influence tool use and decisions.