research-integrated

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to use WebSearch and WebFetch in Step 3 ("웹 조사") to fetch official docs, release notes, CVE databases and other public web content which the agent must read and incorporate into its integrated analysis and recommendations, meaning untrusted third‑party content can influence decisions and next actions.