researcher

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly requires ingesting public user-generated content—e.g., SKILL.md's "Audience" section instructs the agent to "Find real quotes from reviews, forums, social posts" and the deliverables/checklist require screenshots and quotes from competitors—so untrusted third‑party content will be fetched and directly influence positioning and tool decisions.