power-automate-debug

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly calls the FlowStudio MCP server (https://mcp.flowstudio.app) and instructs the agent to fetch and parse action outputs (e.g., outputs.body returned by get_live_flow_run_action_outputs), which can contain arbitrary third-party/user-generated API or webhook responses that the agent reads and uses to decide fixes and next actions, enabling indirect prompt injection.