character-relationships

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

• Third-party content exposure detected (high risk: 0.80). The skill's input requirements state it must ingest a "完整故事文本" provided for analysis (输入要求: 包含人物关系信息的完整故事文本), meaning the agent will read arbitrary user-provided/untrusted story content as part of its workflow, which could carry indirect prompt-injection payloads.