Fly.io

Active

An AI agent for Fly.io. It works in a scene with a live connection to Fly.io. It asks before it writes, and every run is traced.

Manage Fly.io apps and machines. Deploy any Docker image — Bun, Python/uv, Rust, C/clang, or custom containers — to global edge locations. Scale across regions, start/stop machines, and monitor infrastructure.

Infrastructure API Key 4 asset types

What you can connect

Add these to your scene and AI gets access.

Account Accounts

A connected Fly.io account

App Apps

A Fly.io application

Machine Machines

A Fly.io machine (VM instance)

Volume Volumes

Persistent storage volume

Connect Fly.io over MCP

Daslab is itself an MCP server.

Connect Fly.io in a scene, then pin that scene's URL in Claude Code, Cursor, or any MCP client. The agent there gets your Fly.io tools, and your approval gate and trace travel with them.

https://daslab.run/mcp/<your-world>/<your-scene>

One scene can hold Fly.io and everything else you connect, so a single endpoint carries them all. Set up a client.

Use cases

Deploy containerized apps to global edge locations
Manage machine lifecycle — create, start, stop, destroy
Scale apps across multiple regions

From Fly.io

API reference →

Similar integrations

Others in Infrastructure.

SSH Sandbox

Ready to try Fly.io with Daslab?

Get started with CLI