creating-characters

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

• Third-party content exposure detected (high risk: 0.70). The skill's templates (see PROMPT_TEMPLATE.md, especially the "ゆるキャラ調" sections and notes like "生成時に参考画像を添付" / "※添付画像のキャラクターデザイン・動物の種類・色合い・特徴・装飾品に忠実に従うこと") require the agent to ingest and follow user-supplied reference images, which are untrusted third-party content the agent will read/interpret as part of its workflow.