Run Luau scripts headlessly in a Roblox place via Open Cloud for automated testing, game configuration, and procedural generation, with full DataModel and engine Luau API access. Tasks can run for up to 5 minutes, with a limit of 10 concurrent tasks per place. You can save changes using [`SavePlaceAsync`](/reference/engine/classes/AssetService#SavePlaceAsync), pass in large binary payloads, and use [`SerializationService`](/reference/engine/classes/SerializationService) to deserialize RBXM content. For a reference implementation, see [Roblox/place-ci-cd-demo](https://github.com/Roblox/place-ci-cd-demo) on GitHub.

Workflow:

1. [Create a task](#Cloud_CreateLuauExecutionSessionTask__Using_Universes) with your universeId, placeId, and the Luau script to run
2. A server launches, loads the place, and executes your code
3. [Poll for completion](#Cloud_GetLuauExecutionSessionTask) and receive the script's return values and [logs](#Cloud_ListLuauExecutionSessionTaskLogs)

Common use cases include:

- Automated testing: Run test scripts against specific place versions as part of a CI/CD pipeline
- Game configuration: Update pricing, stats, and balancing variables outside of Studio


Represents a dynamically typed value which can be either null, a number, a string, a boolean, a recursive struct value, or a list of values.

A list of LuauExecutionSessionTaskLogs in the parent collection.

The LuauExecutionSessionTaskLogs from the specified
LuauExecutionSessionTask.

A token that you can send as a `pageToken` parameter to retrieve the next
page. If this field is omitted, there are no subsequent pages.

A `LuauExecutionSessionTask` ("task" for short) executes a given Luau script
in the context of a specific version of a place.

In a task, physics simulation does not run. Server and local scripts within
the place also do not automatically run.

The script may access and update the data model of the place, including
invoking any module scripts. However, data model changes are local to the
task and cannot be persisted.

The script can also invoke engine APIs that read and/or modify data stored in
the cloud, such as those for DataStores. Exercise caution when using these
APIs.

Scripts can be up to 4 MB in size and run for up to 5 minutes. Scripts that
run for longer than the time limit terminate with an error.

Scripts are executed as-is and do not need to be wrapped in a function.

Scripts can return values (using the Luau `return` keyword). Return values
are serialized to JSON and can be retrieved with the `Get
LuauExecutionSessionTask` API after the task completes. The total size of the
return values after JSON serialization must not exceed 4 MB. If the limit is
exceeded, the task terminates with an error.

If the script raises an unhandled error, the task terminates. The error
information can be retrieved with the `GetLuauExecutionSessionTask` API.

Standard output (generated by the Luau `print` function) can be retrieved
with the `ListLuauExecutionSessionTaskLogs` method after the task completes.
A maximum of 450 KB of logs are retained. If the amount of logs exceeds the
limit, older logs are discarded.

Information about a task is retained for 24 hours after task completion.

At most ten incomplete tasks are allowed per place. Attempting to create more
tasks while the first ten are incomplete results in a HTTP 429 response.

Resource path of the binary input to this task. See the documentation for
the
`LuauExecutionSessionTaskBinaryInput` resource for usage details.

URI for the binary output of this task. See the `enableBinaryOutput` field
for usage details.

If set to true, allows the task to output a large binary object in addition
to standard return values.

If `enable_binary_output` is set to true, the task script must return a
`LuauExecutionTaskOutput` (or equivalent table) and no other return values.

Below is example code for doing so:

```luau
local buf: buffer = buffer.create(10)
local result: LuauExecutionTaskOutput = {
  BinaryOutput = buf,
  ReturnValues = { "hello world", 123 }
}
return result
```

The `BinaryOutput` buffer must be no larger than 256 MiB in size.

The `ReturnValues` array, if given, will be serialized to JSON and made
available in the `output` field of the `LuauExecutionSessionTask` resource,
similar to regular return values when not using `enable_binary_output`.

The binary output can be fetched from the URI in the `binaryOutputUri`
field after the task completes. The `binaryOutputUri` is valid for 15
minutes after task completion.

The resource path of the luau execution session task.

Formats:
* `universes/{universe_id}/places/{place_id}/luau-execution-session-tasks/{luau_execution_session_task_id}`
* `universes/{universe_id}/places/{place_id}/versions/{place_version_id}/luau-execution-session-tasks/{luau_execution_session_task_id}`
* `universes/{universe_id}/places/{place_id}/luau-execution-sessions/{luau_execution_session_id}/tasks/{luau_execution_session_task_id}`
* `universes/{universe_id}/places/{place_id}/versions/{place_version_id}/luau-execution-sessions/{luau_execution_session_id}/tasks/{luau_execution_session_task_id}`

The script to be run as part of this task.

For example:

```luau
local x = 3
local y = 4
return x + y
```

The task's state. See the State enum for information about each possible
value.

Possible values:

  | Value | Description |
  | --- | --- |
  | STATE_UNSPECIFIED | UNSPECIFIED |
  | QUEUED | The Task is waiting to be processed. |
  | PROCESSING | The Task has been picked up for processing. |
  | CANCELLED | The Task has been stopped by the user. |
  | COMPLETE | The Task has finished processing. The output field contains the output. |
  | FAILED | The Task failed. The error field contains details about the error. |

Limit for how long the script can run.

The task fails if the script does not complete within the
specified duration.

Defaults to 5 minutes.

Time when this task's state last changed.

The user that created the API key that was used to create this task.

Represents a large binary input that can be given to a
`LuauExecutionSessionTask`.

Each `LuauExecutionSessionTaskBinaryInput` is associated with a presigned URL
which can be used to upload an arbitrary object. After uploading the object,
the path of the `LuauExecutionSessionTaskBinaryInput` can be passed when
creating a `LuauExecutionSessionTask` to make the object available to the
task.

Within the task, the contents of the object are made available within a table
passed as the first argument to the script, with the key `BinaryInput`. The
following example demonstrates how to retrieve the data:

```
local taskInput: LuauExecutionTaskInput = ({...})[1]
local buf: buffer = taskInput.BinaryInput
```

Each `LuauExecutionSessionTaskBinaryInput` is valid for 15 minutes from the
time of creation. Within that time, in desired, you can use it to create
multiple tasks for the same universe.

The uploaded binary object must be no larger than 100 MiB.

The resource path of the luau execution session task binary input.

Format:
`universes/{universe_id}/luau-execution-session-task-binary-inputs/{luau_execution_session_task_binary_input_id}`

The size of the binary input object to be uploaded.

The maximum allowed size is 100 MiB.

Upload the binary input object using this URI.

Represents a chunk of log messages generated by a
`LuauExecutionSessionTask`.

The amount of logs that is retained is limited. See the documentation of the
`LuauExecutionSessionTask` resource for more details.

Logs have the same retention time as the parent task.

A list of log messages generated by the task execution.
Each call to the Luau `print()` function from the task's script's code
results in a single list item here, even if the message itself contains
newlines. This field is populated when the View is FLAT (which is the
default).

The resource path of the luau execution session task log.

Formats:
* `universes/{universe_id}/places/{place_id}/luau-execution-session-tasks/{luau_execution_session_task_id}/logs/{luau_execution_session_task_log_id}`
* `universes/{universe_id}/places/{place_id}/versions/{place_version_id}/luau-execution-session-tasks/{luau_execution_session_task_id}/logs/{luau_execution_session_task_log_id}`
* `universes/{universe_id}/places/{place_id}/luau-execution-sessions/{luau_execution_session_id}/tasks/{luau_execution_session_task_id}/logs/{luau_execution_session_task_log_id}`
* `universes/{universe_id}/places/{place_id}/versions/{place_version_id}/luau-execution-sessions/{luau_execution_session_id}/tasks/{luau_execution_session_task_id}/logs/{luau_execution_session_task_log_id}`

A list of log messages generated by the task execution and associated
metadata.
Each call to the Luau `print()` function from the task's script's code
results in a single list item here, even if the message itself contains
newlines. This field is populated when the View is STRUCTURED.

A single log message along with its associated metadata

The time at which the log message was produced

The `MessageType` of the log message.

Possible values:

  | Value | Description |
  | --- | --- |
  | MESSAGE_TYPE_UNSPECIFIED | UNSPECIFIED |
  | OUTPUT | Corresponds to the `MessageOutput` enum value of the engine API. |
  | INFO | Corresponds to the `MessageInfo` enum value of the engine API. |
  | WARNING | Corresponds to the `MessageWarning` enum value of the engine API. |
  | ERROR | Corresponds to the `MessageError` enum value of the engine API. |

Contains error details in case of failed execution.

An error code indicating the category of the error.

Possible values:

  | Value | Description |
  | --- | --- |
  | ERROR_CODE_UNSPECIFIED | UNSPECIFIED |
  | SCRIPT_ERROR | The task's script raised an unhandled error. |
  | DEADLINE_EXCEEDED | The task exceeded the maximum allowed execution duration. |
  | OUTPUT_SIZE_LIMIT_EXCEEDED | The task's outputs were too large. |
  | INTERNAL_ERROR | The Task failed due to an internal system error. |

An error message containing more details about the error.

Contains the output of a task's execution.

Return values from the script that was run. Return values that are not
JSON serializable (such as Data Model Instances) will be returned as
nulls.

A configurable key that allows granular access to Roblox resources. See https://create.roblox.com/docs/cloud/auth/api-keys for more information.

x-api-key

A browser cookie that represents the identity of a Roblox user. DO NOT SHARE THIS. Sharing this will allow someone to log in as you and to steal your Robux and items. We do not recommend using cookies to call endpoints. When possible, use API keys with the x-api-key header or OAuth 2.0 instead for authentication.

.ROBLOSECURITY

Build or authorize apps to access Roblox resources. See https://create.roblox.com/docs/cloud/auth/oauth2-overview for more information.

The Roblox API. See https://create.roblox.com/docs/cloud/open-cloud for more details.

Roblox API

Create a new `LuauExecutionSessionTaskBinaryInput`.

universe_id

Create Luau Execution Session Task Binary Input

LuauExecutionSessionTaskBinaryInput

universe.place.luau-execution-session:write

Creates a task but does not wait for the task to complete.

To check whether a task has
completed, call the `GetLuauExecutionSessionTask` method and inspect the
`state` field of the returned resource.

Quotas:
* 5 calls per minute per API key owner
* 45 calls per minute per IP address

place_id

LuauExecutionSessionTask

version_id

Gets information about a task.

Quotas:
* 45 calls per minute per API key owner or IP address

luau_execution_session_id

task_id

The view in which to retrieve the luau execution session task.

Supports BASIC and FULL.

Defaults to BASIC.

Possible values:

  | Value | Description |
  | --- | --- |
  | VIEW_UNSPECIFIED | The luau execution session task view is not specified; the default will be used. |
  | BASIC | Excludes the `script` field. |
  | FULL | Includes all fields. |

view

universe.place.luau-execution-session:read

Lists log chunks generated by a `LuauExecutionSessionTask`.

Quotas:
* 45 calls per minute per API key owner or IP address

The maximum number of luau execution session task logs to return. The
service might return fewer than this value. If unspecified, at most 10000
luau execution session task logs are returned. The maximum value is 10000
and higher values are set to 10000.

maxPageSize

A page token, received from a previous call, to retrieve a subsequent page.

When paginating, all other parameters provided to the subsequent call must
match the call that provided the page token.

pageToken

The view in which to retrieve the luau execution session task log.

Supports FLAT and STRUCTURED.

Defaults to FLAT.

Possible values:

  | Value | Description |
  | --- | --- |
  | VIEW_UNSPECIFIED | The luau execution session task log view is not specified; the default will be used. |
  | FLAT | If this view is selected, the `messages` field will be populated (and the `structuredMessages` field will not). Each entry of the `messages` array contains only the log message, without additional medata. This is the default. |
  | STRUCTURED | If this view is selected, the `structuredMessages` field will be populated (and the `messages` field will not). Each entry of the `structuredMessages` array contains the log message plus additional metadata (see `LogMessage` for details). |

LuauExecutionSessionTaskLog

Luau Execution

Experiences

Avatar

Assets

Samples and projects

Engine API

Open Cloud API

Studio

Analytics

Platform

Tutorials

Creator programs

Education

Courses

Learn

This method is asynchronous and returns an `Operation`, which represents the current state of the operation. It contains an endpoint path that you can poll to obtain the real response and applicable metadata. See the [documentation]({longRunningOperationsLink}) on long-running operations for more information.
            
The following sections describe the response and metadata objects that are provided when an operation completes.

Beta endpoints are ready for public use. Breaking changes are possible but rare.

These Open Cloud documentation pages will be removed on March 31, 2026. You will be automatically redirected to the new Open Cloud documentation at that time. {linkStart}Learn more{linkEnd}

These old Open Cloud pages will be removed and redirected on March 31, 2026. {linkStart}Learn more{linkEnd}

Experimental endpoints are ready for public use, but may experience breaking changes occasionally as the API contract is finalized.

This endpoint can be called in-experience through {httpServiceLinkStart}HttpService{httpServiceLinkEnd} with an API key.

This field may be set when creating a resource, but may not be modified by subsequent requests.

These fields may be set when creating a resource, but may not be modified by subsequent requests.

This field may be provided in requests, but will never be included in responses.

Legacy authentication method. Always prefer using API Key or OAuth 2.0 when available.

Per API Key: {maxInPeriod} requests every {periodInSeconds} seconds per IP Address

Requires no authentication to call. Typically has significantly lowered rate limits. Always prefer using API Key or OAuth 2.0 when available.

For each grouping, only one of the following fields in the schema has to be provided, the remaining can be excluded.

You can opt back in at any time. {linkStart}Provide feedback.{linkEnd}

You are viewing the old docs. {linkStart}Provide feedback.{linkEnd}

You can opt out at any time and be redirected back to the old reference pages. {linkStart}Provide feedback.{linkEnd}

You are viewing the revamped docs. {linkStart}Provide feedback.{linkEnd}

This field may be set on responses, but will be ignored in requests.

Note: Rate limits can be lower during high-traffic periods. Certain endpoints have additional rate limits. [Learn more about rate limits]({link}).

{limit} requests per day across all API keys for a user or group

{limit} requests per hour across all API keys for a user or group

{limit} requests per minute across all API keys for a user or group

{limit} requests per second across all API keys for a user or group

{limit} requests per minute per API key owner

{limit} request(s) per second per API key owner

Scopes control what permissions your credentials have. Your credentials can only access the endpoints supported by the scopes you selected.

Stable endpoints are production-ready and will almost never experience breaking changes during their lifetime.

This target enables you to limit your key's access to selected {requiredParts}

This target enables you to limit your key's access to selected {requiredParts} and {optionalParts}

Target types are used to restrict the resources that your credentials have permissions on. Only the selected resources can be acted on even if your account has permissions on additional resources. Some scopes do not have target parts. In these cases, all of your relevant resources can be acted on.

The universe Top Level Part does not need to be selected for user-owned API Keys. In this case, your API Key will have access to all of your user-owned experiences and group-owned experiences that you have the right group permissions on.

Try It Out is not yet supported for endpoints requiring API key or OAuth 2.0 authentication

Cookie authentication is disabled for high risk operations for safety.

Critical risk endpoints are extremely sensitive and Try It Out is disabled for safety

Note: Authentication headers are not included in this code sample. You may need to add an authentication header for this command to work.

This is a high risk endpoint that can modify or delete critical data. Use extreme caution when testing.

This API endpoint has been classified as HIGH RISK and may affect live data or user accounts.

Authenticated requests to high-risk APIs can make changes that cannot be undone. Please review your request parameters carefully before proceeding.

Please login to use cookie authentication

This is a low risk endpoint that retrieves data. This is generally safe to test.

This is a medium risk endpoint that modifies data. Test carefully with non-critical accounts, as changes may be permanent.

To keep your account safe, never paste text here from other sources or send information from this page to others.

You are about to test live API endpoints using your actual account credentials.

Only proceed if you understand the implications of the API calls you're about to make. Review the API documentation carefully before testing.

API calls will be made using your authenticated session

These are real API calls that may affect your account or data

Rate limits and other API restrictions apply

Some operations may modify or delete data

This endpoint has been superseded by these recommended alternative endpoints:

The following scopes are required to call this endpoint. For more
information on generating proper keys, see [Manage API keys]({apiKeyReferenceDocLink}).

For more information on generating keys, see [Manage API keys]({apiKeyReferenceDocLink}).

The following endpoints are available at paths relative to the base URL.

This string is formatted as a [FieldMask](/cloud/reference/types#fieldmask).

A JSON value can be `null`, `boolean`, `string`, `number`, `array`, or `object`.

The following scopes are required for your OAuth 2.0 application to use this endpoint. For more information on how to register an OAuth 2.0 application, see [App registration and review]({oauth2ReferenceDocLink}).

For more information on how to register an OAuth 2.0 application, see [App registration and review]({oauth2ReferenceDocLink}).

The following objects describe payloads that are accepted or returned. See each individual endpoint for more information on when these objects are used.

The following scopes are required to call this endpoint.

Resource Paths

This string is formatted as a [Timestamp](/cloud/reference/types#timestamp).

The following scopes are required for your OAuth 2.0 application to use this endpoint. For more information on how to register an OAuth 2.0 application, see [App Registration and Review]({oauth2ReferenceDocLink}).

Not Recommended

You are viewing the revamped Open Cloud reference.

You've opted out of the new Open Cloud reference.

View your ad campaigns and their performance.

Create, update, and manage your ad campaigns.

This allows viewing information about your assets.

This allows uploading and updating assets to Roblox.

This allows viewing asset permissions for subjects (including groups and experiences) and assets your account has access to.

This allows granting asset permissions to any subjects (including groups and experiences) and assets your account has access to.

This allows you to view the status and results of avatar auto-setup jobs on your behalf.

This allows you to create and manage avatar auto-setup jobs on your behalf.

Grants the ability to update badges owned by the user.

This allows viewing information about your commerce items.

This allows making changes to your synced commerce items.

This allows viewing information about your commerce merchant connections.

This allows editing of your commerce merchant connections.

This allows you to view information about your Creator Store products.

This allows you to modify information about your Creator Store products.

This allows you to view information about your Creator Store saves.

This allows you to modify information about your Creator Store saves.

Grants access to create data stores on your behalf.

Grants access to delete data stores on your behalf.

Grants access to read a list of data stores.

Grants access to take a snapshot of your data stores.

Grants access to create entries in your data stores.

Grants access to mark specified entries for deletion.

Grants access to list your data store's entries.

Grants access to read specified entries in your data stores.

Grants access to update the value, attributes, and users of an entry.

Grants access to delete revisions of your data store's entries.

Grants access to list revisions for your data store's entries.

Grants access to read revisions of your data store's entries.

Grants access to view configuration details for the authorized user's developer products.

Grants access to edit the authorized user's developer products.

This allows viewing information about your universe's environments.

This allows editing of your universe's environments.

Grants access to view configuration details for the authorized user's game passes.

Grants access to edit the authorized user's game passes.

Grants access to view community forums information for your account.

Grants access to manage community forums information for your account.

This allows viewing information about your communities.

This allows you to manage community join requests and role updates of lower-ranked community members for your account.

Allows you to view an instance's property data.

Allows you to modify the property data of instances (currently only supports Scripts).

This allows viewing information about your inventory.

Grants access to manage your badges and spend Robux.

Grants access to update the authorized user's badges.

Grants access to manage your developer products.

Grants access to read information about the experiences you have notifications on for.

Grants access to turn on or off notifications for experiences on your behalf.

Grants access to manage your communities.

Grants permission to read community information.

Grants access to read and write your localization tables

Grants access to manage team collaboration for all experiences you own.

Grants access to manage your experiences and certain information associated with your experiences (e.g. localization information).

Grants access to read your experiences' localization information.

Grants access to modify your experiences' localization preferences.

Grants access to manage your user account.

Grants access to modify user localization preferences.

Allows you to view the status, logs, and results of Luau execution tasks for the associated place.

Allows you to create new Luau execution tasks for the associated place.

Grants access to asynchronously flush all data structures in the universe.

Grants access to discard read items from the front of the queue.

Grants access to read items at the front of the queue.

Grants access to read your memory store items.

Grants access to create, update, and delete map items.

This grants the ability to publish messages on behalf of users.

This allows sending notifications from your experience to users.

(Deprecated) Grants access to read a list of entries in an ordered data store.

(Deprecated) Grants access to create a new entry in an ordered data store.

Grants access to list entries from an ordered data store.

Grants access to create entries in an ordered data store.

This allows listing secrets associated with your experience.

This allows creating and updating secrets associated with your experience.

This permission allows you to submit an evaluation job and view the job result.

Grants access to read about your subscription products.

This allows you to read thumbnail information on your behalf.

Grants the ability to read game analytics.

Grants the ability to read place information.

Grants the ability to edit place information.

Grants access to publish new versions of experiences to Roblox.

Grants the ability to read experience information.

Grants access to edit the authorized user's experiences.

Allows you to view user restrictions in the experience.

Allows you to edit user restrictions in the experience.

This allows viewing notifications sent to your account.

Grants the ability to read the authorized user's premium and verified status.

This allows you to view the authorized user's social accounts.

This enables Single Sign-On functionality.

This grants access to read your profile information.

Luau Execution

Places