Configure deployment and skill settings

task brazil Updated Sep 10, 2026 4 min read configure-skill-settings.md
Raw file servicenow.com

Configure where a skill appears in AI Admin Hub, review general information, set security controls, choose a provider, and add evaluation metrics.

Before you begin

Role required: sn_skill_builder.admin

About this task

The Deployment and skill settings tab contains five sections that you can configure in any order:

  • Deployment settings: Choose where the admin can find and activate the skill in AI Admin Hub.
  • General information: Review or edit the skill name and description.
  • Security controls: Restrict which roles can run the skill.
  • Providers: View the language model providers available for the skill.
  • Evaluation metrics: Add metrics that measure the quality of skill responses.

Procedure

  1. Navigate to All > AI Skill Kit > Home.

  2. Select the skill that you want to configure.

  3. Select the Deployment and skill settings tab.

  4. Configure the Deployment settings section.

    [Omitted image "nask-deploy-settings.png"] Alt text: Deployment settings section for AI Skill Kit.

    Under Now Assist features, fill in the fields.

Field

Description

Workflow

High-level category for the skill, such as Technology, Employee, Creator, or Platform. You can also select Other if none of the categories fit.

The workflow that you choose is where the skill appears in the AI Admin Hub console.

Product

Product that the skill operates within, such as ITSM, ITOM, HR Service Delivery, or AI Admin Hub.

Feature

Feature that the skill is used in, such as Agent Chat, Knowledge, or Virtual Agent. You can also define a custom feature.

Name

Name of the feature.

Description

Description of the feature.

Under **Select where you'd like to let the admin activate the skill**, select one or more activation points.
-   **Now Assist panel**
-   **UI Action**
-   **Flow action**
-   **Now Assist context menu**
-   **Virtual assistants**

    For more information about Now Assist in Virtual Agent, see [Configure assistants overview](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/conversational-interfaces/configure-now-assist-va.md).

-   **UI Builder**
  1. Review the General information section.

    This section shows the information that you added when you created the skill. You can edit the skill name and description here.

  2. Configure the Security controls section to restrict which roles can run the skill.

    For the procedure, see Configure security controls for a skill.

  3. Review the Providers section.

    The Providers section lists the language model providers available for the skill. Each provider is tagged as Default or Not supported where applicable.

    Important: The runtime model depends on the configuration in AI Admin Hub and may differ from what's shown in the Providers section.

  4. Add one or more Evaluation metrics to measure the quality of skill responses.

    Evaluation metrics attached in this section apply to every prompt test that you run on the skill.

    Important:

    LLM-judged evaluation metrics consume additional Now Assist assists during prompt testing. Metric execution is billed as a custom call at one assist per 1,000 output tokens, so metrics that produce longer output consume more assists.

    Script-based metrics (metrics with a script instead of a judge prompt) don't consume assists.

    For details about Now Assist consumption during prompt testing, see Test a prompt.

    1. In the Evaluation metrics section, select the add icon [Omitted image "icon-nask-add.png"] Alt text:.

    2. From the list of available metrics, select the metric that you want to add.

      Metrics are grouped by category. Each metric shows the language model provider that runs the metric, such as Amazon Bedrock, Now LLM Generic, or Multiple LLMs. Metrics that run as scripts don't display a provider tag.

      For descriptions of the available metrics, see Evaluate a prompt.

    3. Review the metric details.

      Tab Description
      About Description of the metric, how it works, when to use it, and its output format.
      Judge Prompt / Script Prompt that a language model uses to judge the response, or script that evaluates the response, depending on how the metric is implemented.
    4. Select Add.

  5. Select Save.

What to do next

After you configure the deployment and skill settings, you can publish your skill. To learn more about publishing skills, see Finalize and publish a skill.

Parent Topic:Configuring AI Skill Kit

Related topics

Configure a skill prompt

Configure security controls for a skill

Search documentation
Search document titles and file paths on the brazil branch