orchestrator

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

• Third-party content exposure detected (high risk: 0.90). The skill's instructions explicitly require fetching public third‑party content—references/agent-selection.md maps "Web検索・外部情報収集" to a general-purpose agent and references/examples.md (Example 5) shows a workflow using "WebFetch" to obtain "各社のAPI仕様（生データ）" which the orchestrator ingests and uses to drive comparison and next actions, so untrusted web content can materially influence behavior.