How to Use the Coze MCP in OpenAI Agents SDK

Q: Can my OpenAI Agents SDK system handle multi-agent handoffs to different Coze bots?

Yes. Your agent can call listbots to find active bots in a workspace, then use createchat to delegate specific tasks to them. The SDK manages these interactions as structured tool calls within your existing agent graph.

Q: Can I manage raw file uploads directly from my agent?

You can. Your agent uses uploadfileurl to send external files straight to Coze's storage. It can also run listdatasets to find the right knowledge base target before initiating the upload.

Control Coze bots directly from your OpenAI Agents SDK pipeline with strict execution guardrails and automatic MCP Server discovery.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Coze MCP to OpenAI Agents SDK

Create your Vinkius account to connect Coze to OpenAI Agents SDK and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup Coze with OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Control Coze bots with OpenAI Agents SDK

If you're building with OpenAI Agents SDK, you're likely managing complex handoffs. Exposing the Coze MCP Server tools `create_chat` and `submit_tool_outputs` to your agent lets you delegate tasks to external bots without breaking a sweat. The SDK automatically registers these tools, so your agent knows exactly when to hand off a task to a specific Coze bot. Let's be real: you can't have agents throwing random inputs at production workspaces. The OpenAI Agents SDK validates every payload before calling `publish_bot` or triggering a chat, keeping your Coze setup clean.

Sync knowledge bases on the fly

Keep your OpenAI Agents SDK context window clean. Instead of stuffing raw files into OpenAI, use `upload_document` and `delete_document` to manage your Coze knowledge bases dynamically. Your agent analyzes incoming data and decides what needs to be indexed in Coze. This offloads heavy retrieval tasks to Coze's specialized vector storage. Your OpenAI Agents SDK system runs faster because it only manages the high-level coordination, leaving the document heavy-lifting to `list_datasets` on Coze.

Trace conversation history in the dashboard

Debugging agent-to-bot chats gets messy fast. This integration lets your OpenAI Agents SDK agent pull full chat logs via `get_conversation_history` and clear stale Coze sessions with `clear_conversation`. Every single tool call shows up in your OpenAI dashboard. When your OpenAI Agents SDK agent calls `list_workspaces` to locate a Coze team directory, you can trace the exact payload and response latency in your run logs.

Setup guide

Set up Coze MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Coze tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Coze tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Coze tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Coze Agent",
            instructions="You have access to Coze tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Coze. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Coze now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Coze MCP in OpenAI Agents SDK

Install openai-agents, then instantiate MCPServerStreamableHttp pointing to the Vinkius endpoint. Pass this server instance in the mcp_servers list when constructing your agent. The SDK auto-discovers all eleven tools instantly.

Yes. Your agent can call `list_bots` to find active bots in a workspace, then use `create_chat` to delegate specific tasks to them. The SDK manages these interactions as structured tool calls within your existing agent graph.

Yes. Set cacheToolsList=True in your server configuration. This prevents the SDK from refetching the schemas for tools like `upload_file_url` on every run, reducing startup latency.

You can. Your agent uses `upload_file_url` to send external files straight to Coze's storage. It can also run `list_datasets` to find the right knowledge base target before initiating the upload.

All payloads containing chat messages or raw documents are processed in ephemeral V8 isolates. Vinkius does not store your Coze workspace tokens or the content retrieved via `get_conversation_history`. Your data is encrypted in transit and discarded immediately after the tool execution completes.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript