superlog-debug

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

• Third-party content exposure detected (high risk: 0.75). At runtime, the skill calls Superlog MCP tools like superlog.query_logs / superlog.query_traces, which ingest and return outsider-authored free text from the user’s production telemetry (e.g., log bodies and exception messages emitted by external clients/services), and that returned prose is then placed into the agent’s LLM context.