---
sourceDocument: Yokohama Enable AI
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/yokohama/intelligent-experiences

 Release :

    - yokohama

ft:locale :

    - en-US

ft:publication_title :

    - Yokohama Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Example use case for AI Skill Kit

# Example use case for AI Skill Kit {#ariaid-title1}

Release version: Yokohama  
Updated July 31, 2025  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 2 minutes to read
Summarize  
![AI sparkle icon](https://servicenow.com/docs/portal-asset/ai-sparkle-icon) Summarized using AI  
This content was generated using new OpenAI-powered functionality. Results are provided on an as is basis and are not guaranteed to be accurate or complete.  

## Summary of Example use case for AI Skill Kit

The AI Skill Kit enables developers to create custom AI-driven skills tailored to specific business needs.
This example illustrates how to build a custom skill for summarizing child incidents related to a parent incident, helping ServiceNow customers efficiently organize and understand multiple related incidents.
Show full answer Show less  

## Creating and Configuring the Skill

* Navigate to **All \> AI Skill Kit \> Home** and select **Create new skill**.
* Fill in the form with details such as:
  * Skill name: *Child incident summarization*
  * Description: *Summarization of child incidents*
  * Default provider: *Now LLM Generic*
  * API: *Now LLM Generic*
* After creation, configure the skill settings by selecting a model (e.g., *llmgeneric*) and setting a temperature value (e.g., 0.2) to control output creativity.

## Developing the Prompt

* Create a prompt designed to summarize child incidents by highlighting key issues, impacts, patterns, recurring problems, and outliers.
* Use inputs such as the parent incident's short description and description as well as child incident details, inserted into the prompt with variables like:
  * {{incident.shortdescription}}
  * {{incident.description}}
  * {{ChildIncidents.output}}
* Select appropriate tools such as **Flow Action** and the **IncidentDetailsFetcher** flow resource to support data retrieval for the skill.

## Configuring Deployment Options

* In the Skill settings, enable deployment options to make the skill accessible in the AI Admin Hub.
* Optionally enable UI Action to allow admins to activate the skill directly from the Core UI.

## Testing, Publishing, and Activation

* Run tests by selecting records or incidents to validate that the skill correctly pulls and summarizes data.
* Review the grounded prompt tab to verify input data accuracy.
* Once validated, finalize and publish the prompt to prepare it for production use.
* An admin must activate the published skill in the AI Admin Hub to make it operational.

This process empowers ServiceNow customers to leverage AI for incident management by summarizing related child incidents efficiently, improving incident understanding and resolution workflows.  
As an AI developer, you can create custom skills with AI Skill Kit. For this example, create a custom skill for child incident summarization.

As an AI developer, you must create a skill for child incident summarization. Creating this skill can help you organize and understand multiple incidents that are related to the same parent.

## Create and configure the skill {#example-na-skill-kit__section_csg_fmk_mcc}

To create the skill:

1. Navigate to AllAI Skill KitHome.
2. Select Create new skill.
3. On the form, fill in the fields. The following information is used to fill in the form for this example:
   * Skill name: <kbd class="ph userinput">Child incident summarization</kbd>
   * Description: <kbd class="ph userinput">Summarization of child incidents</kbd>
   * Default provider: <kbd class="ph userinput">Now LLM Generic</kbd>
   * Provider API: <kbd class="ph userinput">Now LLM Generic</kbd>

   {#example-na-skill-kit__ul_uy4_dtf_ncc}

4. Select Create skill.
{#example-na-skill-kit__ol_sx1_xsf_ncc}  
After you create the skill, you must configure the skill settings.

1. Select Configurations.
2. Select the model that you want to use. For this example, you can select <kbd class="ph userinput">llm_generic</kbd>.
3. Select a temperature between 0-1 to determine the randomness and creativity of the output, such as <kbd class="ph userinput">0.2</kbd>.
{#example-na-skill-kit__ol_dzf_h4p_ncc}

## Develop the prompt {#example-na-skill-kit__section_epc_3mk_mcc}

After you create the skill and configure the settings, you must develop the prompt. To follow this example, you can use the following prompt:

You are a customer service representative. Summarize the child incidents of the below given parent incident. The summary should contain key issues and impact across the child incidents, highlighting any patterns, recurring problems, or significant outliers. When summarizing, please consider the following: Parent incident short description: {{incident.short_description}} Parent incident description: {{incident.description}} Here are the child incident details: {{ChildIncidents.output}}

Select the Inputs for the skill. For this example, the record is selected as the input.

Select insert inputs for the prompt.

For this example, the following is used:

<kbd class="ph userinput">Parent incident short description: {{incident.short_description}} Parent incident description: {{incident.description}} Here are the child incident details: {{ChildIncidents.output}}</kbd>

Select the tools for the skill. For this example, select Flow Action and the IncidentDetailsFetcher flow resource.

## Configure the skill deployment options {#example-na-skill-kit__section_imc_nmk_mcc}

The next step is to configure the skill deployment options. These options enable you to choose where to find the skill in AI Admin Hub.

1. Select the Skill settings tab.
2. Select Deployment Settings.
3. Enable the admin to enable the skill from the Core UI by selecting the UI Action check box.
4. Select Save.
{#example-na-skill-kit__ol_cgz_cpp_ncc}

## Test and publish the skill {#example-na-skill-kit__section_xcl_4mk_mcc}

It is important to test your skill prompt to ensure that the correct type of data is being pulled in.

1. Select Run tests.
2. Choose a record or incident.
3. Select Run test.
{#example-na-skill-kit__ol_fnm_4pp_ncc}

Look at the response.

To see the data that was brought into the prompt from your skill inputs and tools, you can look at the grounded prompt tab.

If everything looks good, select Finalize prompt. After you finalize the prompt and you are ready to implement it, select Publish.

## Activate the skill {#example-na-skill-kit__section_epv_qmk_mcc}

After you test, finalize, and publish your skill, an admin must activate it in AI Admin Hub. To learn more about activating skills, see [Activate a skill](https://servicenow-prod.fluidtopics.net/WqOZ2s0eX644czKjibKzNA "After you create and publish a custom skill, you must activate it in AI Admin Hub. Activating the skill enables you to trigger the skill within the UI.").

*[\>]: and then