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

Q: How do I add LangGraph Cloud (Stateful AI Agents) to my Pydantic AI setup?

Install pydantic-ai-slimmcp via pip. Initialize an MCPToolset pointing to your HTTP endpoint. Pass this toolset into your Agent constructor. Do not use the deprecated MCPServerHTTP method.

Q: Will Pydantic AI catch missing variables from LangGraph Cloud (Stateful AI Agents)?

Yes. When your agent calls getthreadstate, the framework validates the incoming JSON against your schema. If a required graph variable is missing, it throws a validation error immediately.

Q: Can I list scheduled jobs using LangGraph Cloud (Stateful AI Agents) in Pydantic AI?

Your agent can call listcrons to see active scheduled jobs. This prevents your system from manually triggering a run that is already queued by a cron schedule.

Q: How do I fix a broken graph state using Pydantic AI?

You pause the execution and use updatethreadstate. This tool allows you to manually override the corrupted graph variables with a validated payload. Once corrected, the agent resumes work.

Integrate LangGraph Cloud with Pydantic AI. Enforce strict type validation on thread states, assistant runs, and graph variables.

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 Pydantic AI

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

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-Safe State Retrieval

The `get_thread_state` tool fetches the current messages and structured outputs so Pydantic AI can validate them instantly. Pulling state from a graph requires knowing exactly what data structure you are getting back. The framework checks the incoming payload against your models. If the graph variables mutate unexpectedly, your agent fails loudly instead of processing corrupt data. You can manually fix these validation errors by injecting corrected data using the `update_thread_state` tool through your MCP connection.

Manage Runs via MCP Server

Triggering an execution is handled strictly by the `create_run` tool, which passes a validated input payload to the thread. The assistant starts processing, and you track the exact progress using `get_run`. You control the entire execution lifecycle from your Python code. We expose `list_runs` so your system can audit every execution attempt on a specific thread. If a process violates your business logic during execution, `cancel_run` kills the job before it completes.

Map Threads and Deployments

The `list_assistants` tool returns your deployed LangGraph configurations before you attempt to run any workflows. Your code maps these directly to specific user intents. This guarantees you never call an outdated or missing graph. Creating a new conversation requires `create_thread`. You can audit your entire active directory using `list_threads`, while `list_crons` gives you visibility into any automated jobs scheduled to run in the background.

Setup guide

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

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "langgraph-cloud-stateful-ai-agents-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to LangGraph Cloud (Stateful AI Agents) tools.",
)

result = await agent.run("List recent LangGraph Cloud (Stateful AI Agents) transactions")
print(result.output)

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 Pydantic AI

Install `pydantic-ai-slim[mcp]` via pip. Initialize an `MCPToolset` pointing to your HTTP endpoint. Pass this toolset into your Agent constructor. Do not use the deprecated `MCPServerHTTP` method.

Yes. When your agent calls `get_thread_state`, the framework validates the incoming JSON against your schema. If a required graph variable is missing, it throws a validation error immediately.

Your agent can call `list_crons` to see active scheduled jobs. This prevents your system from manually triggering a run that is already queued by a cron schedule.

You pause the execution and use `update_thread_state`. This tool allows you to manually override the corrupted graph variables with a validated payload. Once corrected, the agent resumes work.

Every request routes through a V8 Isolate Sandbox on Vinkius. Your thread states, structured outputs, and graph variables are processed ephemerally. The MCP Server requires a single endpoint token for auth and drops all connection data the moment the request finishes.

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