Save infrastructure costs using AI Center APIs!

In Nov 2022, we announced the availability of APIs to programmatically interact with every component within AI Center. You can now manage datasets, train ML models, and deploy ML Skills, all without having to use the UI.

One of the key use-cases for leveraging APIs would be to manage the deployment status of an ML Skill. Depending on the automation, you might not require an ML Skill to be deployed and running all the time. Undeploying ML Skills when not in use could help you save on hardware costs (AI units). In this post, we walk you through an example workflow on how this can be done.

Here are the steps to accessing AI Center APIs. The first two steps enforce access control by making sure your application is registered and authenticated.

  1. Registering an external application (one-time setup)
  2. Generating an authentication token (renew upon expiry)
  3. Making AI Center API calls

Registering an external application

To use these APIs, your app will need to register as an external application with an AI Center scope.

  1. Using the left navigation panel, go to the ‘Admin’ page and select ‘External Applications’.

  2. To add a new external application, click ‘+ Add Application’ and give it a name. Only confidential application type can access AI Center APIs

  3. To enable AI Center scope to the external application, click ‘Add scope’, pick AI Center API Access (preview) from the Resource dropdown, navigate to the Application Scope tab, and select the APIs to which you want to enable access to. Here I’ve selected all APIs for simplicity, but you can choose to assign specific APIs access to specific external applications.

  4. Once you click ‘Add’, a pop up with App ID and an App Secret is generated. You can view the App ID and App Secret anytime from the ‘Edit External application’. These credentials will be used to generate an authentication token in the next step.

Additional information on managing external applications can be found here

Generating an authentication token

For authentication and secure access, the external application must request an access token from the UiPath identity server. The generated auth token expires in one hour, upon which you will need to regenerate a new auth token.

For simplicity, I’ll use Postman in this tutorial. You can register for free, and use their web interface or application. Feel free to use any other tool that you are comfortable with.

Generate a POST request to the Identity Server endpoint that includes the following in the body of the request. Set grant_type to “client_credentials”

client_id: App ID from previous step
client_secret: App Secret from previous step
grant_type: “client_credentials”

If you’re using Postman, make sure you use the content type application/x-www-form-urlencoded.

This request would generate an auth token similar to the one below, also listing the permissions enabled for this external application. We will use this auth token when making AI Center API calls.

Additional information on credentials and auth tokens can be found here.

Making AI Center API calls

The API calls adhere to the following structure where “accountname" and “tenantname” are specific to your account. Most APIs will need the following information

  1. Account ID and Tenant ID: The account and tenant information can be found in the profile info section within the tenant. Read more about it here

  2. Authorization: The authorization field should be populated with the auth token generated in the previous step. When using the auth token, prefix it with ‘Bearer ’ as seen in the screenshot below.

Now we’re ready to access specific APIs. Let’s use sampleClassifierSkill throughout this example. From the UI, you can see this skill exists and is currently in the Available state.

We would be using AI Center APIs to do the following -

  1. Find the ML Skill ID of sampleClassifierSkill
  2. Stop (undeploy) the ML Skill
  3. Start (redeploy) and verify status update.

Find ML Skill details

Using the GET ML Skills call, you can fetch a list of all ML Skills within the account and tenant. You can also add the ML Skill name as part of the query to search for a specific ML Skill and retrieve its details. In this case, we find the ML Skill idfor the sampleClassifierSkill

GET /ai-deployer/v1/mlskills?name=sampleClassifierSkill

In the response below, the key fields are highlighted in yellow. We double check that the sampleClassifierSkill is in the Available state. The ML Skill id will be used in the subsequent APIs to update the state.

Stop ML Skill

To stop an ML Skill

  • Request query should contain update action (‘stop’) and ML Skill id
  • Header should contain account, tenant and auth info

PUT /ai-deployer/v2/mlskills/stop/{mlSkillId}

The response shows the ML skill status change was successful.

Checking the UI, you can verify that sampleClassifierSkill has been undeployed.

Resume ML Skill

To restart the ML Skill (deploy), we make the API calls with the updateType set to Resume

POST /ai-deployer/v1/mlskills/{mlSkillId}?updateType=RESUME

From the response, we see that the request was successful, and the skill state has been updated to ‘Deploying’

Now you can wait for the Skill to become available, and check ML Skill status using the GET ML Skill details API.

From the response, the ML Skill is now deployed and Available! You can also verify the skill availability in the UI. Now your automations can leverage the deployed ML Skill sampleClassifierSkill within workflows to make predictions.

With these API requests, you can programmatically change the state of an ML Skill. These APIs could be used even within an automation workflow to deploy and undeploy an MLSkill, optimizing the uptime for infrastructure and saving costs. While this is just an example, the complete set of AI Center APIs are documented in swagger

where “accountname" and “tenantname” are again specific to your account.

We hope this helps you get started using AI Center APIs, and we look forward to your feedback.

Additional resources

Official documentation


i’ve been using your guide and been able to obtain the access token, though i couldn’t understand how to get the refresh token so it will be actually useful.
would appreciate any support on this :slight_smile:

Hi Daniela, here’s documentation on the refresh token


Hi, can you please tell me how to add additional parameters to update the skill with a specified CPU and Memory? I have tried the below but receive error “{ “respCode”: 60011, “respMsg”: “Resource limits configured incorrectly. Configure both values and make sure limit is greater than or equal to requested values”}”

I am not sure what I could be doing wrong.

When I perform the RESUME API call, the skill does indeed become available, however it always defaults to 0.5 CPU 4 GM RAM

My expected result is this:


I have tried the following:
“processor”: “CPU”,
“deploymentsRequired”: 1,
“gpuRequired”: 0,
“requestedCPU”: 1.0,
“requestedMemory”: 4,
“limitCPU”: 2.0,
“limitMemory”: 8

Hi @louis.intorre,

Kindly try passing the below body:

“processor”: “CPU”,
“gpuRequired”: 0,
“publicSkill”: true,
“autoUpdate”: false,
“inactivityPeriodInDays”: -1,
“replicas”: 1,
“requestCPU”: 1.0,
“requestMemory”: 4,
“limitCPU”: 1.0,
“limitMemory”: 4

You can try passing “requestCPU” instead of “requestedCPU” and similarly for memory as well. You can keep the required parameters in the body as per your needs.

let us know the outcome.