mcp-data-observability-integration

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

• Third-party content exposure detected (high risk: 0.70). The required workflow instructs the agent to load and use MCP templates (e.g., mcp/spark.mcp.json, mcp/kafka.mcp.json, etc.) to query live observability signals at runtime, which typically involves fetching outsider-authored free text from external systems (e.g., log/task output, metadata strings) into the LLM context via the MCP tool responses.