investigate

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

• Third-party content exposure detected (high risk: 0.85). The required workflow repeatedly calls outsider-facing MCP tools (e.g., query-logs, query-traces, query-error-groups, query-sessions) that return free-form message/stack/user/session replay text from external systems, which the agent then incorporates into its LLM context for analysis.