How to Use the Fly.io Extended MCP in Pydantic AI

Q: How do I configure the Fly.io Extended MCP Server with Pydantic AI?

Install the framework using pip install "pydantic-ai-slimmcp" and initialize an MCPToolset pointing to your Vinkius HTTP endpoint. Pass this toolset into the toolsets list when constructing your Pydantic AI Agent to begin managing Fly.io resources.

Q: How does Pydantic AI handle long-running operations like waiting for a Fly.io Extended machine to start?

The server includes the waitmachine tool, which blocks execution until a machine reaches a specified state. Your Pydantic AI agent calls this tool within its execution loop, receiving a validated state response once the machine is fully operational.

Q: Can I run the Fly.io Extended MCP Server locally with Pydantic AI?

Yes, though the unified MCPToolset in Pydantic AI requires the MCP server to be running externally. You can host the server on Vinkius and connect via the secure SSE or Streamable HTTP transport options provided in your dashboard.

Manage Fly.io apps and volumes with type-safe runtime validation using Pydantic AI and the Fly.io Extended MCP Server.

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 Extended MCP to Pydantic AI

Create your Vinkius account to connect Fly.io Extended 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 Fly.io Extended with Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Type-Safe Provisioning with Pydantic AI

`create_app` lets your Pydantic AI agent initialize Fly.io applications with strict runtime schema validation. If the Fly.io API returns unexpected configuration fields, Pydantic AI raises a validation error immediately, preventing corrupt state from creeping into your infrastructure pipeline. This MCP Server integrates with Pydantic AI's unified `MCPToolset` to expose clean schemas for `create_machine`. The agent can confidently spin up machines using precise configuration objects, knowing that each parameter is validated against strict Python type hints before execution.

Validated Fly.io Volume Management

`create_volume` provides your Pydantic AI agent with a reliable tool for provisioning persistent storage on Fly.io. Because Pydantic AI validates all tool outputs at runtime, your agent receives structured, type-safe data when calling `get_volume` or `list_volumes`. This strict validation prevents the agent from hallucinating volume IDs or capacity metrics. If you need to scale storage, the agent executes `extend_volume` with verified dimensions, ensuring your database nodes expand without silent failures or configuration drift.

Deterministic Machine Control

`start_machine` gives your Pydantic AI agent direct, type-safe control over Fly.io compute lifecycles. The agent manages active workloads by calling `stop_machine` or `suspend_machine` based on real-time application demands, with all state changes validated against Pydantic models. By using `wait_machine`, your agent halts execution until the target machine achieves the exact state expected. This eliminates race conditions during rolling updates, providing SRE teams with a highly deterministic and predictable automation layer.

Setup guide

Set up Fly.io Extended 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": {
        "flyio-extended-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to Fly.io Extended tools.",
)

result = await agent.run("List recent Fly.io Extended 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 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 Extended 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 Extended MCP in Pydantic AI

Install the framework using `pip install "pydantic-ai-slim[mcp]"` and initialize an `MCPToolset` pointing to your Vinkius HTTP endpoint. Pass this toolset into the `toolsets` list when constructing your Pydantic AI `Agent` to begin managing Fly.io resources.

Pydantic AI enforces strict runtime validation on all MCP server tool outputs. This means commands like `get_machine` or `list_apps` return strongly typed data, forcing the agent to fail loudly if Fly.io's API returns unexpected payloads, preventing silent infrastructure corruption.

The server includes the `wait_machine` tool, which blocks execution until a machine reaches a specified state. Your Pydantic AI agent calls this tool within its execution loop, receiving a validated state response once the machine is fully operational.

Yes, though the unified `MCPToolset` in Pydantic AI requires the MCP server to be running externally. You can host the server on Vinkius and connect via the secure SSE or Streamable HTTP transport options provided in your dashboard.

Your Fly.io API tokens and sensitive machine configurations are isolated within Vinkius's secure, zero-trust V8 sandboxed environment. The MCP server processes your requests in memory without writing credentials to disk, ensuring your infrastructure access keys remain private.

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