compliant-position

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md only describes two runtime tool calls (gbrain search, then potentially web_search), and if web_search is executed it would fetch public web content at runtime and ingest the resulting page text into the agent’s LLM context (outsider-authored free text via public web content).