How to Use the Coze MCP in LangChain

Q: Can I trace Coze MCP Server calls in LangChain using LangSmith?

Yes. Every time your LangChain agent calls tools like createchat or listbots through this MCP Server, LangSmith logs the exact inputs, outputs, and latency.

Q: How do I manage Coze session memory inside a LangChain runnable?

You fetch the message log using getconversationhistory and feed it into your LangChain state. To wipe the session clean for a fresh run, call clearconversation.

Q: How do I configure this Coze MCP Server inside my LangChain codebase?

Install langchain-mcp-adapters and use MultiServerMCPClient pointing to the Vinkius endpoint. Then, pull the tool definitions and pass them directly to your LangChain agent creator.

Q: Does this MCP Server expose my Coze chat history and uploaded documents to third parties?

No. Your Coze chat history from getconversationhistory and text payloads in uploaddocument pass through an isolated V8 sandbox on Vinkius. No data is stored or shared; it goes straight to your agent.

Run multi-step LangChain pipelines that spin up, configure, and chat with Coze bots on the fly.

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 LangChain

Create your Vinkius account to connect Coze to LangChain 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 LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Coze bot execution in LangChain pipelines

This MCP Server lets you connect LangChain agents directly to Coze workspaces to deploy and trigger bots as part of a larger chain. The `publish_bot` tool lets your LangChain agent deploy a draft bot, while `create_chat` sends messages and handles the response loop immediately. This setup lets you build sequential chains where a LangChain agent first checks available workspaces using `list_workspaces` and then routes the output to a specific Coze bot. You get full observability over these Coze tool calls via LangSmith tracing.

Manage RAG datasets within LangChain chains

This MCP Server lets your LangChain agent feed raw data directly into Coze knowledge bases. Your LangChain agent uses `upload_document` to push text files or `upload_file_url` to ingest external links, making them instantly searchable for the Coze bot. If a document becomes stale during a chain run, the LangChain agent invokes `delete_document` to purge it from Coze. This keeps your external data syncs clean without manual dashboard intervention.

ReAct agents handling Coze tool outputs

This MCP Server lets LangChain ReAct agents use `submit_tool_outputs` to feed execution results back to a running Coze bot chat. When a Coze bot requires a local plugin execution, your LangChain agent handles the local run and submits the final output. You can also use `get_conversation_history` to pull past Coze messages into LangChain memory before starting a new chat session. If the context gets too heavy, the LangChain agent calls `clear_conversation` to reset the state.

Setup guide

Set up Coze MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes Coze tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "coze-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent Coze transactions"
    })
    print(result["messages"][-1].content)

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 LangChain

Your LangChain agent runs local code or APIs to resolve what the bot needs, then uses `submit_tool_outputs` to send the results back. This keeps the execution loop entirely inside your LangChain chain.

Yes. Every time your LangChain agent calls tools like `create_chat` or `list_bots` through this MCP Server, LangSmith logs the exact inputs, outputs, and latency.

You fetch the message log using `get_conversation_history` and feed it into your LangChain state. To wipe the session clean for a fresh run, call `clear_conversation`.

Install `langchain-mcp-adapters` and use `MultiServerMCPClient` pointing to the Vinkius endpoint. Then, pull the tool definitions and pass them directly to your LangChain agent creator.

No. Your Coze chat history from `get_conversation_history` and text payloads in `upload_document` pass through an isolated V8 sandbox on Vinkius. No data is stored or shared; it goes straight to your agent.

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