Integration

Daytona integration

Daytona sandbox platform API: create, list, start, stop, archive and delete the isolated sandboxes that run AI-generated code, snapshot a running sandbox or build a snapshot from an image, manage the persistent volumes sandboxes mount, label sandboxes so they can be found again, and resolve the preview URL of a port or the toolbox base URL of a sandbox. Commands and file operations inside a sandbox are served by a separate per-sandbox toolbox host and are not part of this API.

Connect Daytona How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Daytona.Built from the 18 Daytona tools below, on the connection you add once.

Common Daytona use cases

cleaning up obsolete environment templates
choosing a snapshot to create a sandbox from
keeping a dataset across disposable sandboxes
opening a dev server started inside a sandbox
finding a sandbox created earlier by its labels
tagging a sandbox with the workflow run that owns it
reclaiming storage once a dataset is no longer needed
stopping a sandbox to stop paying for it between runs
waiting until a new sandbox reaches the started state
cleaning up a disposable sandbox at the end of a workflow

What you can do with Daytona

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Sandbox, 8 operations

archive_sandboxArchive sandboxPOSTMoves a stopped sandbox to archive storage while keeping its filesystem.
create_sandboxCreate sandboxPOSTCreates an isolated sandbox with a chosen size, image, volumes and lifecycle.
delete_sandboxDelete sandboxDELETEDeletes a sandbox permanently, discarding its filesystem.
get_sandboxGet sandboxGETReads a sandbox with its state, resources, volumes and toolbox URL.
list_sandboxesList sandboxesGETLists sandboxes filtered by state, labels, snapshot, region or size.
set_sandbox_labelsSet sandbox labelsPUTReplaces every label on a sandbox with the supplied map.
start_sandboxStart sandboxPOSTStarts a sandbox and returns it with its new desired state.
stop_sandboxStop sandboxPOSTStops a running sandbox while keeping its filesystem.

Snapshot, 5 operations

create_sandbox_snapshotCreate sandbox snapshotPOSTCaptures the current state of a sandbox as a snapshot new sandboxes can start from.
create_snapshotCreate snapshotPOSTRegisters a snapshot built from a container image, with resource defaults.
delete_snapshotDelete snapshotDELETEDeletes a snapshot so no new sandbox can be created from it.
get_snapshotGet snapshotGETReads a snapshot with its build state and resource defaults.
list_snapshotsList snapshotsGETLists snapshots with their state, image and resource defaults.

Volume, 3 operations

create_volumeCreate volumePOSTCreates a persistent volume that sandboxes can mount and share.
delete_volumeDelete volumeDELETEDeletes a persistent volume and everything stored on it.
list_volumesList volumesGETLists persistent volumes with their state and last use.

Preview url, 1 operation

get_port_preview_urlGet port preview urlGETReturns the browser-reachable URL and access token for a sandbox port.

Toolbox url, 1 operation

get_toolbox_proxy_urlGet toolbox proxy urlGETReturns the base URL of the toolbox host that serves in-sandbox operations.

Use Daytona from Claude Code, Cursor or Codex

Connect Daytona in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Daytona operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Daytona

Create your workspace
Sign up free. Nothing to install.
Connect Daytona once
Add your Daytona connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Daytona integration FAQ

What can you do with the Daytona integration?

LiveContext exposes 18 operations on Daytona, covering sandbox, snapshot, volume and preview url. Common uses include cleaning up obsolete environment templates, choosing a snapshot to create a sandbox from and keeping a dataset across disposable sandboxes.