How to Use the Fly.io MCP in OpenAI Agents SDK

Q: How does Fly.io machine creation work within the OpenAI Agents SDK?

Your agent calls createmachine through this MCP integration to deploy Firecracker VMs on the Fly.io edge. The OpenAI Agents SDK automatically registers this tool, allowing the model to spin up instances dynamically based on incoming traffic demands.

Q: Can I run diagnostics on a Fly.io VM using OpenAI Agents SDK tools?

Yes, you can run commands like ls or database migrations using execmachine directly from your agent. The OpenAI Agents SDK traces the exact shell output, giving you full visibility of the remote execution inside your OpenAI dashboard.

Q: How does the OpenAI Agents SDK prevent accidental Fly.io volume deletion?

The agent uses listvolumes to check for attached NVMe storage before executing any teardown commands. Because the OpenAI Agents SDK supports safety guardrails, you can block the agent from calling deletemachine if an active volume is detected.

Spin up, scale, and debug Fly.io edge machines safely using our MCP Server and the OpenAI Agents SDK.

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 OpenAI Agents SDK

Create your Vinkius account to connect Fly.io to OpenAI Agents SDK 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 OpenAI Agents SDK

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Safe Edge Scaling via OpenAI Agents SDK

The `create_machine` tool provisions physical Firecracker VMs directly from your OpenAI Agents SDK pipeline. When your traffic spikes, the agent detects the load and boots instances in specific edge regions without human intervention. By calling `delete_machine`, the agent tears down idle VMs once demand drops, avoiding runaway compute bills. Built-in guardrails in this Python SDK validate each scaling call against your budget limits before spinning up new hardware.

Automated VM Recovery and Diagnostics

The `get_machine` tool pulls live execution states and image digests straight into your OpenAI Agents SDK runtime. If a worker node crashes or suspends, the agent reads this state and executes `start_machine` to bring it back online. For deeper issues, the agent triggers `exec_machine` to run diagnostic shell commands inside the active VM. The SDK traces these execution logs directly on your OpenAI dashboard, so you see exactly what command the agent ran and why.

Stateful Volume and App Management

The `list_volumes` tool identifies active NVMe storage drives attached to your application before your agent attempts any data-destructive actions. This prevents the OpenAI Agents SDK from deleting critical database volumes when scaling down compute. Your agent coordinates these storage operations using our hosted MCP server to check the network topology with `get_app` and listing active spaces via `list_apps`. You get isolated execution environments for every run, keeping your infrastructure changes deterministic and safe.

Setup guide

Set up Fly.io MCP in OpenAI Agents SDK

Prerequisites

Python 3.10+ installed
openai-agents package (pip install openai-agents)
Active Vinkius subscription with a valid endpoint token

1

Install the SDK

Run pip install openai-agents to install the OpenAI Agents SDK. The MCP integration is built-in — no extra dependencies needed.
2

Connect via SSE transport

Use MCPServerSse with your Vinkius endpoint URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. The SDK auto-discovers all Fly.io tools at runtime.
3

Create your Agent

Pass the MCP to Agent(mcp_servers=[server]). The agent receives Fly.io tools as native definitions — JSON schemas resolve automatically.
4

Run the agent

Call Runner.run(agent, prompt) to execute. The agent invokes the appropriate Fly.io tools and returns structured results. Copy the full example on the right to get started.

agent.py

import asyncio
from agents import Agent, Runner
from agents.mcp import MCPServerSse

async def main():
    async with MCPServerSse(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    ) as server:
        agent = Agent(
            name="Fly.io Agent",
            instructions="You have access to Fly.io tools.",
            mcp_servers=[server],
        )
        result = await Runner.run(agent, "List recent transactions")
        print(result.final_output)

asyncio.run(main())

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 OpenAI Agents SDK

Your agent calls `create_machine` through this MCP integration to deploy Firecracker VMs on the Fly.io edge. The OpenAI Agents SDK automatically registers this tool, allowing the model to spin up instances dynamically based on incoming traffic demands.

Yes, you can run commands like `ls` or database migrations using `exec_machine` directly from your agent. The OpenAI Agents SDK traces the exact shell output, giving you full visibility of the remote execution inside your OpenAI dashboard.

The agent uses `list_volumes` to check for attached NVMe storage before executing any teardown commands. Because the OpenAI Agents SDK supports safety guardrails, you can block the agent from calling `delete_machine` if an active volume is detected.

No, Vinkius handles all the underlying authentication for you. You use a single endpoint token in your SDK configuration, and the platform manages the secure connection to the Fly.io API.

Your Fly.io VM configurations and API tokens run inside a secure V8 Isolate Sandbox on Vinkius. The MCP Server never exposes your raw credentials to the LLM, passing only the stripped JSON outputs of tools like `get_machine` 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