Isolated environments for executing untrusted or AI-generated code in Linux containers or isolated Workers, controlled through the Sandbox SDK's TypeScript API for commands, files and exposed services. GA on the Workers Paid plan since April 2026.
FluffyStack toolsJurisdictional exposure
Sub-services (4)
Sandbox SDK
TypeScript API to run commands, manage files and stream process output
Container sandboxes
Full Linux environments built from a user-provided image
Dynamic Workers
Lightweight isolates that load untrusted JavaScript, Python or Wasm at runtime
Preview URLs
Public URLs for web services running inside a sandbox
Compliance & Certifications
This service is attested for the following frameworks. Always verify with the provider before relying on a specific compliance posture.
Where this runs
Commercial regions (29)
Europe (10)
- Paris
- Frankfurt
- Dublin
- Milan
- Amsterdam
- Warsaw
- Madrid
- Stockholm
- Zurich
- London
North America (4)
- Toronto
- Ashburn
- Chicago
- San Jose
South America (2)
- Buenos Aires
- São Paulo
Asia (6)
- Hong Kong
- Mumbai
- Tokyo
- Singapore
- Seoul
- Taipei
Oceania (2)
- Sydney
- Auckland
Middle East (2)
- Tel Aviv
- Dubai
Africa (3)
- Lagos
- Cape Town
- Johannesburg
Tags
Equivalent services on other platforms
Kubernetes-native sandbox for running untrusted AI-agent code on GKE, using gVisor kernel isolation, with pod snapshots for pause and resume and high-rate sandbox creation. Built on the open-source Kubernetes Agent Sandbox project.
Isolated Firecracker microVMs for running untrusted or agent-generated code, with Linux images, managed or custom OCI images, snapshots, persistent sandboxes and drives; JS/Python SDKs and CLI; billed on active CPU and provisioned memory