workflow-wf

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly instructs the agent to perform autonomous WebSearch/WebFetch of external websites as part of the Research and Prompt workflows (see "Web검색/코드탐색 자율 수행" and Research → "정보 수집" which use WebSearch/WebFetch to collect official docs, blogs, and other web content), and those fetched, potentially user-generated or untrusted sources are summarized and used to drive recommendations and next actions.