How to Use the LangGraph Cloud (Stateful AI Agents) MCP in Mastra AI

Q: How do I handle a failed agent run with LangGraph Cloud (Stateful AI Agents) in Mastra AI?

Your Mastra AI workflow can use getrun to detect a 'failed' status. From there, you can use Mastra's built-in retry logic or branch to a different path, like notifying a human or using updatethreadstate to reset the agent to a known good state.

Q: Can I build a human approval step using LangGraph Cloud (Stateful AI Agents) and Mastra AI?

Absolutely. Your workflow executes a run, then uses getthreadstate to get the agent's proposed action. The workflow pauses, you send the data for human review, and then use updatethreadstate with the approval before continuing.

Q: How do I choose which agent to run in my Mastra AI workflow?

Use the listassistants tool at the start of your workflow. This will return the available agent configurations from LangGraph Cloud. You can then pass the chosen assistant's ID to the createrun tool.

Build durable, stateful agent workflows in Mastra AI. This server connects you to LangGraph's persistent agent memory.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect LangGraph Cloud (Stateful AI Agents) MCP to Mastra AI

Create your Vinkius account to connect LangGraph Cloud (Stateful AI Agents) to Mastra AI 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 LangGraph Cloud (Stateful AI Agents) with Mastra AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Orchestrate Runs as Workflow Steps

Treat LangGraph agent executions as reliable steps in your Mastra AI workflows. The `create_run` tool kicks off a graph execution. It’s not fire-and-forget; you get a run ID back that you can track. Your workflow can then poll `get_run` to check the status. If it's done, move to the next step. If it's stalled, branch to an error-handling path. And if a run needs to be stopped, `cancel_run` provides the immediate off-switch.

Manage State as a Resource

State isn't a side effect; it's the core of your workflow. A Mastra AI job can start by calling `create_thread` to provision a new memory space for an agent. This makes your agent's context explicit and manageable. At any point, you can inspect the agent's brain with `get_thread_state`. For human-in-the-loop tasks, your workflow can pause, present the state to a user, and then use `update_thread_state` to inject the user's decision back into the agent's memory before resuming.

Monitor Your Fleet of Agents

This MCP Server gives you the tools to see everything that's happening. `list_assistants` shows you what agent configurations are even available to run. `list_threads` gives you a manifest of all active conversations. For debugging or auditing, `list_runs` shows you every execution tied to a specific thread. And with `list_crons`, you can verify that your scheduled, automated agent jobs are configured correctly. It's the visibility you need to run autonomous systems in production.

Setup guide

Set up LangGraph Cloud (Stateful AI Agents) MCP in Mastra AI

Prerequisites

Node.js 18+ and a TypeScript project
@mastra/mcp + @mastra/core packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run npm install @mastra/mcp @mastra/core plus your preferred model provider (e.g. @ai-sdk/openai).
2

Configure the MCPClient
Create an MCPClient with your Vinkius endpoint as a URL object. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Discover and inject tools
Call mcpClient.listTools() and spread the result into your agent's tools object. All LangGraph Cloud (Stateful AI Agents) tools become native Mastra tools.
4

Run with any model
Swap openai("gpt-4o") for any AI SDK-compatible provider. Call agent.generate() and the agent routes tool calls through MCP automatically.

agent.ts

import { MCPClient } from "@mastra/mcp";
import { Agent } from "@mastra/core/agent";
import { openai } from "@ai-sdk/openai";

const mcpClient = new MCPClient({
  id: "langgraph-cloud-stateful-ai-agents-mcp-client",
  servers: {
    "langgraph-cloud-stateful-ai-agents-mcp": {
      url: new URL(
        "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
      ),
    },
  },
});

const agent = new Agent({
  name: "LangGraph Cloud (Stateful AI Agents) Agent",
  model: openai("gpt-4o"),
  instructions: "You have access to LangGraph Cloud (Stateful AI Agents) tools.",
  tools: {
    ...(await mcpClient.listTools()),
  },
});

const result = await agent.generate(
  "List recent LangGraph Cloud (Stateful AI Agents) transactions"
);
console.log(result.text);

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 LangGraph Cloud. 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 LangGraph Cloud (Stateful AI Agents) 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 LangGraph Cloud (Stateful AI Agents) MCP in Mastra AI

Your Mastra AI workflow can use `get_run` to detect a 'failed' status. From there, you can use Mastra's built-in retry logic or branch to a different path, like notifying a human or using `update_thread_state` to reset the agent to a known good state.

Absolutely. Your workflow executes a run, then uses `get_thread_state` to get the agent's proposed action. The workflow pauses, you send the data for human review, and then use `update_thread_state` with the approval before continuing.

Use the `list_assistants` tool at the start of your workflow. This will return the available agent configurations from LangGraph Cloud. You can then pass the chosen assistant's ID to the `create_run` tool.

Yes. You can build a step in a monitoring workflow that calls `list_threads`. This will give you the IDs for all active conversation threads, which you can then display or use for health checks.

The server handles LangGraph thread state data, like agent messages and structured outputs. Your MCP endpoint token secures all requests. Your server runs in a sandboxed V8 Isolate, so your data is never mixed with other users'.

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