How to Use the Nhost MCP in LangChain

Q: How do I use the Nhost MCP server with LangChain?

You'll use the langchain-mcp-adapters package. Just point the MultiServerMCPClient at your Vinkius endpoint, call .gettools(), and pass the resulting list directly into your agent definition.

Q: What's the best way to handle file uploads with Nhost in LangChain?

The uploadfile tool is what you need. Your agent can take a file path or byte stream as input, call the tool, and get back the file metadata from Nhost Storage. You can then chain that output into another step.

Build auth and storage workflows into your LangChain agents. Let them sign up users, manage files, and handle sessions in a single chain.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Nhost MCP to LangChain

Create your Vinkius account to connect Nhost 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 Nhost with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Auth Flows from Start to Finish

Your agent can now run a complete user onboarding sequence. It starts by calling `signup_email_password`, then immediately uses `signin_email_password` to get a session token, all within one logical chain. If a session expires, the agent doesn't just fail. It can be designed to automatically use the stored refresh token with `refresh_token` to get a new JWT and retry the original failed operation. That's the whole point of chaining these tools.

Connect Storage Operations to Logic

Go beyond simple auth. A LangChain agent can take user input, call `upload_file` to store it in Nhost, and then pass the resulting file metadata to another tool in the same execution. You can build chains that retrieve data, process it, and clean up afterwards. For example, an agent could fetch a temporary file with `get_file_presigned_url`, use its contents, and then explicitly call `delete_file` when it's done. This MCP connection makes that possible.

Build Smarter LangChain Agents

This isn't just about calling APIs. It's about giving your agent the context to make decisions. It can check a user's status with `get_user` before deciding whether to let them upload a file or change their email. Your agent can handle more complex scenarios, like passwordless flows. It can initiate the process with `signin_passwordless_email` and then wait for the user to click the magic link before continuing its task. This MCP server makes those tools available for your agent's reasoning loop.

Setup guide

Set up Nhost 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 Nhost 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({
    "nhost-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 Nhost 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 Nhost. 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 Nhost 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 Nhost MCP in LangChain

You'll use the `langchain-mcp-adapters` package. Just point the `MultiServerMCPClient` at your Vinkius endpoint, call `.get_tools()`, and pass the resulting list directly into your agent definition.

Yes. While agents are stateless by default, you can use the client's session context manager. This lets the agent hold onto things like a `refresh_token` so it can stay 'logged in' for the next task.

The `upload_file` tool is what you need. Your agent can take a file path or byte stream as input, call the tool, and get back the file metadata from Nhost Storage. You can then chain that output into another step.

Absolutely. Since these are standard LangChain tools, every call—inputs, outputs, and latency—is automatically traced in LangSmith. You can see exactly when your agent decided to call `get_user` or `signout`.

Your agent will handle user credentials like email and password for tools like `signin_email_password`, and it will receive JWTs and refresh tokens in return. All traffic is encrypted over HTTPS and the server runs in an isolated sandbox, but your agent's logic is responsible for how it manages those tokens.

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