AgentApp API Invocation

Once an AgentApp is deployed and listening on 127.0.0.1:8090, you can issue inference calls via the streaming endpoint, the OpenAI-compatible endpoint, and the AG-UI endpoint.

Prerequisites

• The AgentApp is running (the demo agent is named Friday in examples).

• Model credentials (for example DASHSCOPE_API_KEY) are configured in the environment.

• Prefer HTTP clients that support async SSE, such as aiohttp or httpx.

/process: SSE Streaming Endpoint

Request Body

{
  "input": [
    {
      "role": "user",
      "content": [
        { "type": "text", "text": "What is the capital of France?" }
      ]
    }
  ],
  "session_id": "optional, reuse for the same session",
  "user_id": "optional, helps differentiate users"
}

• input follows the Agentscope message schema and can hold multiple messages and rich content.

• session_id is used to enable services like StateService and SessionHistoryService to record context, support multi‑turn memory, and persist the agent’s state.

• user_id defaults to empty; supply it if you need per-account accounting.

Parsing the Stream

The server responds with Server-Sent Events, each line starting with data: {...} and ending with data: [DONE]. The snippet below (taken from test_process_endpoint_stream_async) demonstrates how to parse incremental text:

async with session.post(url, json=payload) as resp:
    assert resp.headers["Content-Type"].startswith("text/event-stream")
    async for chunk, _ in resp.content.iter_chunks():
        if not chunk:
            continue
        line = chunk.decode("utf-8").strip()
        if not line.startswith("data:"):
            continue
        data_str = line[len("data:") :].strip()
        if data_str == "[DONE]":
            break
        event = json.loads(data_str)
        text = event["output"][0]["content"][0]["text"]
        # Accumulate or display text here

Multi-turn Example

The following example shows how to reuse session_id across multiple requests to achieve effects such as “remembering the user’s name.”

1. First call: "My name is Alice."

2. Second call: "What is my name?"

The SSE stream will mention “Alice”, confirming that the session state is in effect.

session_id = "123456"

async with aiohttp.ClientSession() as session:
    payload1 = {
        "input": [
            {
                "role": "user",
                "content": [{"type": "text", "text": "My name is Alice."}],
            },
        ],
        "session_id": session_id,
    }
    async with session.post(url, json=payload1) as resp:
        assert resp.status == 200
        assert resp.headers.get("Content-Type", "").startswith(
            "text/event-stream",
        )
        async for chunk, _ in resp.content.iter_chunks():
            if not chunk:
                continue
            line = chunk.decode("utf-8").strip()
            if (
                line.startswith("data:")
                and line[len("data:") :].strip() == "[DONE]"
            ):
                break

payload2 = {
    "input": [
        {
            "role": "user",
            "content": [{"type": "text", "text": "What is my name?"}],
        },
    ],
    "session_id": session_id,
}

async with aiohttp.ClientSession() as session:
    async with session.post(url, json=payload2) as resp:
        assert resp.status == 200
        assert resp.headers.get("Content-Type", "").startswith(
            "text/event-stream",
        )

        found_name = False

        async for chunk, _ in resp.content.iter_chunks():
            if not chunk:
                continue
            line = chunk.decode("utf-8").strip()
            if line.startswith("data:"):
                data_str = line[len("data:") :].strip()
                if data_str == "[DONE]":
                    break
                try:
                    event = json.loads(data_str)
                except json.JSONDecodeError:
                    continue

                if "output" in event:
                    try:
                        text_content = event["output"][0]["content"][0][
                            "text"
                        ].lower()
                        if "alice" in text_content:
                            found_name = True
                    except Exception:
                        pass

        assert found_name, "Did not find 'Alice' in the second turn output"

/compatible-mode/v1/responses: OpenAI-Compatible Endpoint

If your system already uses the official OpenAI SDK, simply point it to this endpoint for near drop-in compatibility:

from openai import OpenAI

client = OpenAI(base_url="http://127.0.0.1:8090/compatible-mode/v1")
resp = client.responses.create(
    model="any_name",
    input="Who are you?",
)
print(resp.response["output"][0]["content"][0]["text"])

The endpoint reuses the Responses API schema. The test_openai_compatible_mode test also confirms that the model replies with its agent name “Friday”.

/ag-ui: AG-UI Protocol Compatible Endpoint

AgentScope Runtime supports the AG-UI Protocol through an adapter pattern. Agent APIs built with the @app.query decorator are automatically adapted to the AG-UI Protocol request format.

AG-UI Request Body Format

In the HTTP API, the body of POST /ag-ui is a RunAgentInput.

{
  "threadId": "thread_123",
  "runId": "run_456",
  "parentRunId": "run_000",
  "state": {},
  "messages": [
    {
      "id": "msg_1",
      "role": "user",
      "content": "Hello, please introduce yourself"
    }
  ],
  "tools": [],
  "context": [],
  "forwardedProps": {}
}

Field Descriptions

• threadId (required): Conversation thread ID, used to identify a dialogue session

• runId (required): Current run ID, should use a unique ID for each request

• messages (required): Message list, supporting multiple message types:

  • role: "user": User message

  • role: "assistant": Assistant message (can include text or tool calls)

  • role: "system": System message

  • role: "tool": Tool execution result message

• tools (optional): Available tool list

• context (optional): Context information

• state (optional): State information

• forwardedProps (optional): Forwarded properties

API Response

The server returns SSE (Server-Sent Events) format streaming responses, with each event starting with data: and containing JSON format event data. Currently supported event types include:

Event Types

Event Type	Description	Example Fields
run_started	Run started	thread_id, run_id
text_message_start	Text message started	message_id
text_message_content	Text message incremental content	message_id, delta
text_message_end	Text message ended	message_id
tool_call_start	Tool call started	tool_call_id, tool_call_name
tool_call_args	Tool call arguments	tool_call_id, delta
tool_call_end	Tool call ended	tool_call_id
tool_call_result	Tool execution result	tool_call_id, content
run_finished	Run completed	thread_id, run_id
run_error	Run error	run_id, message, code

Usage Example

import aiohttp
import json

async def call_ag_ui_endpoint():
    url = "http://localhost:8080/ag-ui"
    payload = {
        "threadId": "thread_1234",
        "runId": "run_4567",
        "messages": [
            {
                "id": "msg_1",
                "role": "user",
                "content": "What's the weather like in Beijing today?"
            }
        ]
    }

    async with aiohttp.ClientSession() as session:
        async with session.post(url, json=payload) as resp:
            assert resp.status == 200
            assert resp.headers["Content-Type"].startswith("text/event-stream")

            async for chunk, _ in resp.content.iter_chunks():
                if not chunk:
                    continue

                line = chunk.decode("utf-8").strip()
                if not line.startswith("data:"):
                    continue
                print(line + "\n")

Response example:

data: {"type": "RUN_STARTED", "thread_id": "thread_1234", "run_id": "run_4567"}

data: {"type": "TOOL_CALL_START", "tool_call_id": "call_c51915f8d0ab4c6aac85e1", "tool_call_name": "get_weather", "message_id": "msg_57562f7d-e96d-4de9-8899-1334fc101e23_0"}

data: {"type": "TOOL_CALL_ARGS", "tool_call_id": "call_c51915f8d0ab4c6aac85e1", "delta": "{\"location\": \"Beijing\"}"}

data: {"type": "TOOL_CALL_END", "tool_call_id": "call_c51915f8d0ab4c6aac85e1"}

data: {"type": "TOOL_CALL_RESULT", "message_id": "msg_0ca9a23b-0674-496b-91c8-5bd699945e70_0", "tool_call_id": "call_c51915f8d0ab4c6aac85e1", "content": "[{\"type\": \"text\", \"text\": \"The weather in Beijing is sunny with a temperature of 25°C.\"}]", "role": "tool"}

data: {"type": "TEXT_MESSAGE_START", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "role": "assistant"}

data: {"type": "TEXT_MESSAGE_CONTENT", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "delta": "The weather"}

data: {"type": "TEXT_MESSAGE_CONTENT", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "delta": " in Beijing today"}

data: {"type": "TEXT_MESSAGE_CONTENT", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "delta": " is sunny,"}

data: {"type": "TEXT_MESSAGE_CONTENT", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "delta": " with a temperature of 25"}

data: {"type": "TEXT_MESSAGE_CONTENT", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0", "delta": "°C."}

data: {"type": "TEXT_MESSAGE_END", "message_id": "msg_8debb51f-3226-4f1a-a573-5f80db132f80_0"}

data: {"type": "RUN_FINISHED", "thread_id": "thread_1234", "run_id": "run_4567"}

Frontend Integration

Using CopilotKit, you can quickly build Agent applications based on existing AG-UI Protocol APIs. For details, refer to the CopilotKit documentation.

Troubleshooting

• Cannot connect: Ensure the service is running, the port is free, and the client targets the correct host (especially in containers or remote deployments).

• Parsing errors: SSE frames must be processed line by line; tolerate keep-alive blank lines or partial JSON chunks.

• Context not retained: Confirm every request includes the same session_id, and that the state/session services are properly started within init_func.

With the examples above, you can quickly validate and consume a deployed Agent service.