Developer Tools & DevOps

Connect Daytona to Your AI Agents

Enable your Wkil agents to provision sandboxes, run commands, clone repositories, and manage snapshots through a secure Daytona integration.

Authentication: API key

Build with Daytona

What you can do

Provision and manage sandbox environments
Run commands and clone repositories inside sandboxes
Snapshot and reuse sandbox state

Available capabilities may vary by authentication and workspace configuration.

Actions supported by the connector

114 verified tools

Create Sandbox

Spin up a new isolated development sandbox environment.

Execute Command

Run a shell command inside a sandbox.

Create Snapshot

Save a sandbox's current state as a reusable snapshot.

Clone Repository

Clone a Git repository into a sandbox.

Stop Sandbox

Stop a running sandbox to free its resources.

Create Sandbox Backup

Start a backup operation for an existing sandbox.

Archive Sandbox

Archive a sandbox that is no longer actively used.

Create Organization

Create a new organization workspace in Daytona.

Create API Key

Create a new API key for the organization.

View all verified tools (search)

Archive SandboxCancel Organization InvitationCheckout Git BranchClone RepositoryCreate API KeyCreate Sandbox BackupCreate Folder in SandboxCreate OrganizationCreate Organization InvitationCreate Organization RoleCreate Process SessionCreate PTY SessionCreate Docker RegistryCreate SandboxCreate SnapshotCreate SSH AccessCreate VolumeDeactivate SnapshotDelete API KeyDelete API Key for UserDelete Sandbox FileDelete Git BranchDelete OrganizationDelete Organization OTEL ConfigDelete Organization RoleDelete PTY SessionDelete RecordingDelete Docker RegistryDelete SandboxDelete SessionDelete SnapshotDelete VolumeDownload File (Deprecated)Download Multiple FilesEnroll in SMS MFAExecute CommandExpire Signed Port Preview URLFind In Files (Deprecated)Focus Accessibility NodeGet API KeyGet Available Account ProvidersGet Build Logs URLGet Commit HistoryGet Computer Use StatusGet Daytona ConfigGet Display InfoGet Entrypoint LogsGet File Info (Deprecated)Get Git History (Deprecated)Check Sandbox AccessGet Daytona Health StatusGet OrganizationGet Organization Audit LogsGet Organization Invitations CountGet Organization Usage OverviewGet Process StatusGet Project Dir (Deprecated)Get PTY SessionGet Push AccessGet Docker RegistryGet SandboxGet Session Command LogsGet Session (Deprecated)Get SnapshotGet Snapshot Build LogsGet Toolbox Proxy URLGet Transient Push AccessGet Authenticated UserGet User Home Dir (Deprecated)Get VolumeGet WindowsGet Working DirectoryGet Work Dir (Deprecated)List API KeysList Available RegionsList Files (Deprecated)List Git Branches (Deprecated)List Organization InvitationsList Organization Invitations for UserList organization membersList organization rolesList OrganizationsList Process SessionsList PTY SessionsList registriesList SandboxesList Sandboxes PaginatedList shared regionsList SnapshotsList VolumesReplace In FilesReplace Sandbox LabelsResize PTY SessionRevoke SSH AccessSearch FilesSet Sandbox Auto Archive IntervalSet Auto Delete IntervalSet Autostop IntervalStart Computer UseStart LSP ServerStart VNC ProcessStart SandboxStop LSP ServerStop SandboxTake Compressed ScreenshotUnlink AccountUpdate Experimental ConfigUpdate Last ActivityUpdate Organization InvitationUpdate Organization OpenTelemetry ConfigUpdate Organization RoleUpdate Public StatusUpdate Docker RegistryUpdate Sandbox Default Limited Network Egress

Practical agent use cases

An agent spins up a fresh sandbox to test a code change before it is merged.
An agent clones a repository and runs its test suite inside an isolated sandbox.

How it works with Wkil

1
Choose the integration
Add Daytona to your agent from inside the Wkil platform.
2
Grant the permissions it needs
Define precisely what the agent can access and perform.
3
Test, then run it
Test actions before they go live, with human approval where needed.

Permissions always stay in your control — you can scope access and require human approval before any sensitive action.

Related integrations

FAQ

Can an agent run arbitrary commands unattended?

Command execution can be scoped and configured to require approval for anything beyond routine, pre-approved actions.