supervisor-api-client-function-calling

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md's "Mixed Flow: MCP Approval + Client-Side Function Tools" shows a uc_connection tool described as "Searches the web for current information" (and an example input "Search the web for what Databricks is..."), meaning the agent ingests untrusted public web content returned by that hosted tool into its conversation history and uses it to decide follow-up actions.