Can my agent scale up my application by creating more machines in a specific region?

Yes. Use the 'create_machine' tool. You can specify the region (e.g., 'iad', 'gru') and the Docker image. The agent will command the Fly.io orchestrator to boot up new VM instances instantly to handle increased demand.

How do I run a database migration or shell command on a live machine via chat?

Use the 'exec_machine' tool. Provide the machine ID and the command. The agent bypasses SSH entirely by interacting directly with the hypervisor API to run your binary alongside the running container securely.

Can I check the health and image version of my active machines through the agent?

Absolutely. The 'get_machine' tool retrieves exhaustive runtime states, including container health, current executing status (started/stopped), and the precise docker image digest actively booted into memory.

Fly.io MCP Server for Google ADK 10 tools — connect in under 2 minutes

Name: Fly.io
Availability: InStock
Author: Vinkius

google.github.io/adk-docs/tools/mcp-tools

Built by Vinkius GDPR 10 Tools SDK

Google Agent Development Kit (ADK) is Google's framework for building production AI agents. Add Fly.io as an MCP tool provider through the Vinkius and your ADK agents can call every tool with full schema introspection.

Get MCP Server for AI Agents

ASK AI ABOUT THIS MCP SERVER

Open in ChatGPT Open in Claude Open in Perplexity

Vinkius supports streamable HTTP and SSE.

python

from google.adk.agents import Agent
from google.adk.tools.mcp_tool import McpToolset
from google.adk.tools.mcp_tool.mcp_session_manager import (
    StreamableHTTPConnectionParams,
)

# Your Vinkius token — get it at cloud.vinkius.com
mcp_tools = McpToolset(
    connection_params=StreamableHTTPConnectionParams(
        url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    )
)

agent = Agent(
    model="gemini-2.5-pro",
    name="flyio_agent",
    instruction=(
        "You help users interact with Fly.io "
        "using 10 available tools."
    ),
    tools=[mcp_tools],
)

Fully ManagedVinkius Servers

60%Token savings

High SecurityEnterprise-grade

IAMAccess control

EU AI ActCompliant

DLPData protection

V8 IsolateSandboxed

Ed25519Audit chain

<40msKill switch

Stream every event to Splunk, Datadog, or your own webhook in real-time

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure

About Fly.io MCP Server

Connect your Fly.io account to any AI agent and take full control of your edge computing and container orchestration through natural conversation.

Google ADK natively supports Fly.io as an MCP tool provider — declare the Vinkius Edge URL and the framework handles discovery, validation, and execution automatically. Combine 10 tools with Gemini's long-context reasoning for complex multi-tool workflows, with production-ready session management and evaluation built in.

What you can do

App Space Orchestration — List logical parent containers (Apps) spanning the Fly Edge network, identifying anycast assignments and dedicated IPv4/IPv6 IPs natively
Machine Runtime Management — Navigate and control individual MicroVM (Machine) endpoints, fetching unique IDs and explicit placement regions flawlessly
Autonomous Scaling — Provision new highly available Edge Machines to scale horizontal capacities dynamically without waiting on full platform deployments
Live Health Auditing — Examine exhaustive runtime states, returning dynamic executing statuses (started, stopped, suspended) and docker image digests in real-time
Remote Command Execution — Inject and run shell commands inside active Machines bypassing SSH by interacting directly with the hypervisor API securely
Persistent Storage Control — List hardware NVMe Volumes attached to your apps to manage stateful data like PostgreSQL or SQLite independent of compute
Network DNA Extraction — Retrieve the operational baseline of Fly Apps, identifying Wireguard ranges and cluster master regions synchronously

The Fly.io MCP Server exposes 10 tools through the Vinkius. Connect it to Google ADK in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.

How to Connect Fly.io to Google ADK via MCP

Follow these steps to integrate the Fly.io MCP Server with Google ADK.

Install Google ADK

Run pip install google-adk

Replace the token

Replace [YOUR_TOKEN_HERE] with your Vinkius token

Create the agent

Save the code above and integrate into your ADK workflow

Explore tools

The agent will discover 10 tools from Fly.io via MCP

Why Use Google ADK with the Fly.io MCP Server

Google ADK provides unique advantages when paired with Fly.io through the Model Context Protocol.

Google ADK natively supports MCP tool servers — declare a tool provider and the framework handles discovery, validation, and execution

Built on Gemini models, ADK provides long-context reasoning ideal for complex multi-tool workflows with Fly.io

Production-ready features like session management, evaluation, and deployment come built-in — not bolted on

Seamless integration with Google Cloud services means you can combine Fly.io tools with BigQuery, Vertex AI, and Cloud Functions

Fly.io + Google ADK Use Cases

Practical scenarios where Google ADK combined with the Fly.io MCP Server delivers measurable value.

Enterprise data agents: ADK agents query Fly.io and cross-reference results with internal databases for comprehensive analysis

Multi-modal workflows: combine Fly.io tool responses with Gemini's vision and language capabilities in a single agent

Automated compliance checks: schedule ADK agents to query Fly.io regularly and flag policy violations or configuration drift

Internal tool platforms: build self-service agent platforms where teams connect their own MCP servers including Fly.io

Fly.io MCP Tools for Google ADK (10)

These 10 tools become available when you connect Fly.io to Google ADK via MCP:

create_machine

Scales horizontal capacities dynamically without waiting on full platform deployments. Provision a new highly available Edge Machine inside an App

delete_machine

The Firecracker VM is dropped, compute billing ceases immediately, and any ephemeral disk state dissolves. Fails safely if persistent volumes are explicitly attached without the force flag. Terminate and destroy a Fly Machine forever (Scale Down)

exec_machine

Useful for `ls`, `ps aux`, `top`, or running internal database diagnostic migrations. Inject and run a shell/Bash command inside an active Fly Machine

get_app

Identifies the primary Region holding the cluster master, internal Wireguard network ranges assigned, and any active Anycast IPs actively routing inbound user traffic globally. Retrieve the operational baseline state of a distinct Fly App

get_machine

Returns dynamic executing states ("started", "stopped", "suspended"), the precise docker image digest/SHA actively booted into RAM, and any mapped volume points tying persistent SQLite/Postgres logs. Get exhaustive runtime states attached to a single Fly Machine

list_apps

Apps are fundamentally distinct collections of individual microVMs (Machines), dedicated IPv4/IPv6 anycast assignments, and persistent storage volumes. List Fly.io App spaces belonging to an Organization

list_machines

Retrieves unique identifiers and explicit placement Regions (e.g., iad, ams, nrt). List individual MicroVM (Machine) endpoints inside a Fly App

list_volumes

Crucial identifier for managing stateful applications (PostgreSQL, SQLite, persistent cache) safely independent of compute instances. List persistent hardware NVMe Volumes attached to an App

start_machine

Utilized extensively when recovering paused batch processors or restarting crashed worker nodes dynamically across edge points of presence. Boot a previously stopped or suspended Fly Machine

stop_machine

Drastically reduces latency bills during idle cycles outside typical user ingress bands. Gracefully halt a running Fly.io internal Machine

Example Prompts for Fly.io in Google ADK

Ready-to-use prompts you can give your Google ADK agent to start working with Fly.io immediately.

"List all machines in my 'web-api' app"

"Run 'ls -la /app' on machine '918572b0' in app 'web-api'"

"Show me the persistent volumes for 'web-api'"

Troubleshooting Fly.io MCP Server with Google ADK

Common issues when connecting Fly.io to Google ADK through the Vinkius, and how to resolve them.

McpToolset not found

Update: pip install --upgrade google-adk

Fly.io + Google ADK FAQ

Common questions about integrating Fly.io MCP Server with Google ADK.

How does Google ADK connect to MCP servers?

Import the MCP toolset class and pass the server URL. ADK discovers and registers all tools automatically, making them available to your agent's tool-use loop.

Can ADK agents use multiple MCP servers?

Yes. Declare multiple MCP tool providers in your agent configuration. ADK merges all tool schemas and the agent can call tools from any server in a single turn.

Which Gemini models work best with MCP tools?

Gemini 2.0 Flash and Pro models both support function calling required for MCP tools. Flash is recommended for latency-sensitive use cases, Pro for complex reasoning.