This endpoint lists all of the audit logs which your API key has access to.

Auth scope: Audit API Key

Audit Log Model

Audit logs are representations of a specific action taken by a user in the Gamesight platform. Each Audit Log contains a few key properties:

Actor - The user or API key that took the action
Verb - The type of action taken (eg create, update, or delete)
Target - The object that the action was taken on. For example when updating the name on a Team the Target would be the Team that you updated.
Context - The Org, Game, or Team context that this action occurred in. For example when creating a Tracker the Context would be set to the Team that the Tracker was created in.

Below you can find a full specification of the Audit Log model

Column	Type	Nullable	Description
`id`	Number	No	Unique ID for this audit log
`actor_type`	String	No	Type of actor. One of: `user`, `advertiser_api_key`, `system`
`actor_id`	Number	Yes	ID of the actor. Not set when `actor_type` is `system`
`actor_name`	String	Yes	Name for the actor. Either the user's full name or the API key's name
`actor_email`	String	Yes	Email for the actor. Only set for user actors.
`actor_ip`	String	Yes	IP address for the actor. IP is not collected by default. If you would like to include IP addresses in your Audit Logs please contact your account manager to enable this setting
`verb`	String	No	The type of action that was recorded. Once of: `create`, `update`, `delete`
`target_type`	String	No	The type of object that the actor mutated for this audit log
`target_id`	String	No	The ID of the mutated object
`target_meta`	Object	Yes	Additional metadata about the mutated object. Currently populated for `actor_role` and `permission` target types
`attribute`	String	Yes	The name of the attribute that was updated. Only set for `update` verb audit logs
`old_value`	String	Yes	The previous value for the updated attribute. Only set for `update` verb audit logs
`new_value`	String	Yes	The new value for the updated attribute. Only set for `update` verb audit logs
`context_type`	String	No	The type of context that this action was taken in. One of: `org`, `game`, `team`. Can be used to filter audit logs to only include actions that impact a specific Game or Team.
`context_id`	Number	No	The ID for the context that this action was taken in
`created_at`	String	No	ISO 8601 format timestamp in UTC that this action was taken at

Target Types

There are several different target_type values that you may see in your audit logs. Some of these object names may be unfamiliar as they are internal system concepts, so we have defined them here:

refresh_token - A create refresh token audit log entry is generated on the start of every session for a user. You can use this log entry to see user "logins" or session starts
actor_role - Each time a user or API key is assigned a role in your Organization you will see a create actor role entry. This will contain information about which role was assigned to the user and in what context.
permission - In some cases permissions are directly assigned to a user, rather than through a role assignment. In these cases you will see the target type as create permission rather than actor role.

Soft Deletes

There are a number of target_types that implement a soft-delete system to allow for data recovery. In these cases you may see what is presented as a delete action in the product appear in the Audit Log as an update with an attribute of archived_at. The soft delete logic is implemented as:

archived_at set to the deletion timestamp to mark a record for deletion
archived_at set to null to restore a deleted record

Pagination

This endpoint has cursor based pagination.

{
  "audit_logs": [...],
  "cursors": {
    "next": "eyJpZCI6IDM2MjE3fQ=="                
  }
}

To retrieve the next page in the time range look for cursors.next on the response payload. Once there are no additional pages in the time range the next cursor will return null.

Below is an example implementing this cursor based pagination in Python.

audit_logs = []
cursor = None

while True: # Continue pulling pages until there are no more pages
  resp = get_audit_log_page(cursor)
  
  audit_logs.append(resp['audit_logs'])
  cursor = resp['cursors']['next']
  
  if cursor is None: # No more pages, break out of the loop
    break

200200