How to Use the Fly.io MCP in LangChain

Q: How do I set up the Fly.io MCP Server inside a LangChain project?

Install the adapter package using pip install langchain-mcp-adapters langgraph. Initialize the MultiServerMCPClient pointing to your Vinkius endpoint, fetch the tools with client.gettools(), and pass them to your agent constructor.

Q: Can a LangChain agent automatically clean up idle Fly.io machines?

Yes. Your agent can run a background chain that calls listmachines to find idle microVMs. It then triggers stopmachine or deletemachine to stop billing for those specific Fly.io instances.

Deploy and scale Fly.io edge instances dynamically using LangChain chains that link machine state checks to instant scaling decisions.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Fly.io MCP to LangChain

Create your Vinkius account to connect Fly.io 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 Fly.io with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Fly.io scaling decisions with LangChain logic

Build multi-step chains where your agent checks active workloads before spinning up new compute. This setup uses `list_machines` with our MCP Server to inspect your current fleet, feeding that raw data directly into the next link in your LangChain pipeline. Your agent makes the call to scale out only when actual traffic demands it. If the workload spikes, the chain automatically triggers `create_machine` to provision a new Firecracker microVM in the target region. You get full visibility into this entire execution flow through LangSmith tracing, showing you exactly how your agent decided to scale.

Run remote diagnostics through LangChain agents

Let your ReAct agents troubleshoot running instances without manual SSH sessions. The agent invokes `exec_machine` to run diagnostic commands like checking database migrations or system logs directly inside your active Fly.io machines. When a machine fails a health check, the agent reads the error and immediately runs `get_machine` to inspect the image digest. It can then decide to trigger a reboot using `start_machine` or tear down a corrupted instance with `delete_machine` based on the trace history.

Manage stateful volumes using this MCP Server

Combine LangChain memory with physical storage tracking. Your agent checks persistent NVMe storage availability across your cluster by calling `list_volumes` before attaching them to new compute instances. This MCP Server lets your agent coordinate storage and compute safely. The agent verifies which volumes are free, runs `create_machine` with the correct mount points, and ensures your database has its disk ready before booting.

Setup guide

Set up Fly.io 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 Fly.io 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({
    "flyio-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 Fly.io 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 Fly.io. 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 Fly.io 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 Fly.io MCP in LangChain

Install the adapter package using `pip install langchain-mcp-adapters langgraph`. Initialize the `MultiServerMCPClient` pointing to your Vinkius endpoint, fetch the tools with `client.get_tools()`, and pass them to your agent constructor.

Yes. Your agent can run a background chain that calls `list_machines` to find idle microVMs. It then triggers `stop_machine` or `delete_machine` to stop billing for those specific Fly.io instances.

LangSmith logs every single tool call, including the parameters sent to `exec_machine` or `create_machine`. You can inspect the exact payload, latency, and response from the Fly.io API directly in your tracing dashboard.

The tool returns the raw API error to your agent. Your agent can catch this failure in the chain, analyze the issue, and try deploying to an alternative region using `list_apps` to find healthy clusters.

Your Fly.io API tokens and machine configuration parameters run inside a zero-trust V8 Isolate sandbox. Vinkius executes these calls ephemerally, meaning your infrastructure credentials and environment variables are never written to persistent disk or exposed to other tenants.

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