- ISOLATION
- PERSISTENCE
- CLEANUP
- HUMAN
Short answer
How can I use AI to manage test environments?
Your AI uses Daytona to fork existing sandboxes and attach persistent storage for testing. This allows you to create, use, and clean up isolated environments through natural language. You get a repeatable way to test code in fresh, identical setups.
The outcomes of automation.
Where your test data lands.
Once connected, your AI manages the entire lifecycle of your test infrastructure.
ISOLATION
Clean sandbox clones
Your AI creates fresh copies of your standard environments. This ensures every test run starts from a known, clean state.
PERSISTENCE
Managed data volumes
The AI handles storage by creating and attaching volumes. Your test data stays put even when the sandbox is destroyed.
CLEANUP
Automated teardown
Your agent wipes away unused volumes and sandboxes. This prevents storage bloat and keeps your workspace tidy.
HUMAN
Manual oversight
You decide when to keep a volume or delete a sandbox. The AI follows your specific instructions for every environment lifecycle event.
The workflow
What your AI does when the request arrives.
The AI translates your intent into specific infrastructure actions.
Clone the environment
The AI creates a fresh, isolated workspace based on your existing setup.
fork_sandboxSetup storage
Your agent creates a new volume to hold the data needed for the test.
create_volumeVerify data
The AI checks the available storage to ensure the right volume is ready.
list_volumesget_volume_by_nameClean up
After the test finishes, the AI removes the temporary storage.
delete_volume
Try it
Copy these to start.
Use these prompts to trigger environment actions.
Starting points
These are starting points. Swap out the repo names and volume names for your actual project details.
Daytona Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_AkqYHK4XFPqtgrdtnVnobFEfCwP5QQF4omfr3EXo/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Daytona capabilities are ready to use.
{
"mcpServers": {
"daytona-dev-workspaces-mcp": {
"url": "https://edge.vinkius.com/vk_preview_AkqYHK4XFPqtgrdtnVnobFEfCwP5QQF4omfr3EXo/mcp"
}
}
}Fork a new sandbox from my 'production-template' repo.
Create a new volume named 'test-data-vol' for my current workspace.
List all my existing volumes so I can see what is active.
Delete the volume named 'temp-test-storage' once the run is done.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Daytona once, then ask.
Just link your account once. Your credentials stay encrypted, and you can immediately start managing workspaces through your chat interface.
Connect Daytona to your AIFAQ
How this task behaves.
- 01
Can the AI delete my main production database?
No. The AI only has the ability to delete volumes through the specific tool provided. It cannot touch your primary production databases or infrastructure outside of these managed volumes.
- 02
Does the AI automatically delete sandboxes after I'm done?
No. The AI only deletes sandboxes or volumes if you explicitly tell it to do so in your prompt.
- 03
Can I see what volumes are currently active?
Yes. Your AI can list all existing volumes to show you what is currently taking up space.
- 04
Can the AI create a sandbox from any random URL?
No. The AI can only fork sandboxes that are part of your existing Daytona environment setup.
- 05
Will the AI remember my data if I don't create a volume?
No. If you don't use the tool to create a persistent volume, any data written to a sandbox will be lost when that sandbox is removed.
More questions about Daytona? See everything the Daytona Connector can do
Connect Daytona to Claude, Cursor, ChatGPT & more
