Getting Started
This page outlines upcoming features on how to build and run desktop AI agents seamlessly within ToyStack's virtual operating system. These features are currently under development and not yet live.
Deploy your instance
Step-1
Generate API key
Sign up on our dashboard. An API key will be generated for you automatically.
Step-2
Install SDK
Install the Python SDK with pip.
Step-3
Configure client
Configure the client with your API key.
Step-4
Start instance of virtual os
Interact with the Instance
You can interact with your ToyStack Virtual OS instance using various tools such as screenshot
, bash
, computer
, and edit
. Below are examples of how to perform common actions:
Integrate with Claude (Optional)
You can optionally connect Claude Computer Use to your ToyStack Virtual OS instance using BashTool
, ComputerTool
, EditTool
, and ToolResult
.
Note: This is an optional integration. If you are using Claude's Computer Use, ensure you have the correct API keys and configurations.
Stop the Instance
When you are done with your session, it is recommended to stop the instance. This action will delete all data stored during the session.
Upcoming Features
ToyStack is actively developing additional features to support desktop AI agents and enhanced tooling within the Virtual OS. Stay tuned for updates as these features are rolled out in future releases
Last updated