{
  "_meta": {
    "description": "D6 fixtures for ms-agent-python / agent-config",
    "sourceFile": "harness/fixtures/d5/agent-config.json",
    "created": "2026-05-22"
  },
  "fixtures": [
    {
      "match": {
        "userMessage": "tone:professional",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "Greetings. I am operating in professional tone. I will provide measured, formal responses calibrated to your stated preferences and refrain from colloquialisms in my replies."
      }
    },
    {
      "match": {
        "userMessage": "tone:casual",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "Hey! Casual mode here — I'll keep things friendly and easygoing. Just shoot me whatever you want to know and I'll riff on it without sounding like a press release."
      }
    },
    {
      "match": {
        "userMessage": "expertise:beginner",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "Sure! Think of CopilotKit as a friendly toolkit. It helps you add an AI helper to your app. The helper can answer questions, run small tasks, and show buttons or charts when needed."
      }
    },
    {
      "match": {
        "userMessage": "expertise:expert",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "CopilotKit composes a runtime adapter (Express/Hono) over the AG-UI SSE protocol; the React client wires hooks (useFrontendTool, useAgentContext) into a typed agent runner. The architecture front-runs round-trip latency by streaming TEXT_MESSAGE_CHUNK and TOOL_CALL events on the same channel."
      }
    },
    {
      "match": {
        "userMessage": "responseLength:concise",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "Agent context is a typed payload sent each turn."
      }
    },
    {
      "match": {
        "userMessage": "responseLength:detailed",
        "context": "ms-agent-python"
      },
      "response": {
        "content": "Agent context is a typed payload published from the frontend on every turn through the useAgentContext hook. The payload is forwarded into the agent's runtime context (LangGraph 0.6 introduced the `context` channel as the supported relay for per-run frontend-supplied data; legacy `properties` flowed via `forwardedProps` and did not land in `RunnableConfig`). On the Python side, CopilotKitMiddleware reads the value off the runtime context, then routes it into the system-prompt builder so the model sees the user's tone, expertise, and length preferences before each call. The result is per-turn behavior change without a model swap."
      }
    }
  ]
}