Create a Prompt Builder AI Skill

Overview

This topic covers how to create an AI skill and build a prompt to send to generative artificial intelligence (AI) models.

A prompt builder AI skill helps you to create a prompt to send to generative AI models. You'll tell the model what you want it to do using plain language. After adjusting a few settings and testing the prompt, you can see what the model returns based on your instructions.

Here's a high-level breakdown of how to create and use a prompt builder AI skill:

1. Create an AI skill
2. Describe what you want the skill to do
3. Adjust the temperature
4. Provide examples
5. Select a model
6. Enable extended thinking
7. Test the prompt
8. Use the skill!

Create an AI skill

1. In the Build view, click NEW > AI Skill.

   

2. Click TRY PROMPT BUILDER.

   

3. Configure the following properties:

   Property	Description
   Name	Enter a name that follows the recommended naming standard.
   Description	(Optional) Enter a brief description of the AI skill.

4. Click CREATE.
5. On the Review AI Skill Security window, configure security.
6. Click SAVE. The AI skill opens in a new dialog or window.

Describe what you want the skill to do

To start, instruct the model on what to do with the information you'll give it. Use plain language to describe what you're going to send the model and what you expect in response.

For example, you could instruct the model to:

• Summarize long or complex pieces of text
• Generate responses or other content
• Extract certain pieces of information from larger pieces of text

You can also include details about what to exclude, the format of the response, as well as the length or tone.

Not sure what to say? That's okay! You can start by describing the minimum requirements for the model. Then, add some example inputs and outputs. You can test the prompt directly in the AI skill to see if the model responded in a way you need and expect.

What makes a good prompt?

As you get started, remember the model doesn't have any context or awareness of what your end goal is. The model isn't like an old coworker who knows your habits or expectations. Instead, the model is more like a brand new hire who's learning and eager to help. You'll have to be very clear in your prompt, describing your task and your expectations. This is why examples are helpful, too.

Let's use an example to compare a good prompt to a bad prompt, and how the results differ. Imagine we're going to use this model to summarize form submissions on our website. Customers sometimes submit support requests through the general inquiry form, and we want to make sure that we respond to them in a timely manner. We get hundreds of submissions a day, which takes a lot of time to manually review and triage. So our business is building an Appian process to summarize the form submissions and involve a human being when the model spots a support inquiry.

Based on that use case, consider how these prompts might perform:

Bad prompt	Good prompt
Summarize the key information in this form and let me know if it's a support request.	From the content of this form submission, answer the following questions: What type of submission is this? What information is important in this submission? What team would be responsible for responding to this submission?

When the model analyses a form submission from a customer asking for assistance with their electric bike, here are the responses it returns:

Bad prompt response

Good prompt response

This appears to be a support request regarding a persistent battery issue affecting the reliability of an electric bike. The customer has attached images of error messages and is seeking clear instructions or steps to promptly address the malfunctioning battery.

This is a customer support submission. The important information in this submission includes: The customer is reporting a persistent issue with their electric bike battery malfunctioning since January 1st customer has followed troubleshooting steps in the manual but the problem persists The customer has attached images of error messages shown on the bike's console The customer is seeking guidance and clear instructions to resolve the battery malfunction promptly The team responsible for responding to this submission would be the electric bikes customer support/service team, as the customer is reporting an issue with their electric bike battery and is seeking assistance from the company.

Specifying the output format is especially helpful if you plan to use the response data elsewhere in your process. You can ask the model to return that information in the format you need, and you won't need to transform it using a script task in the process.

Adjust the temperature

Next, choose the temperature of the prompt. This value tells the model how controlled or creative to be with its answers.

As a generative AI model creates a response, it uses probability to determine the word that most likely comes next in a sequence. Temperature controls the probability that the model uses to choose the next word in its response.

A lower temperature means that the model should only choose words where there's a high probability it comes next in a sequence. With a lower temperature, you can expect more consistent answers each time the model is prompted.

Conversely, a higher temperature tells the model that it can choose words with a lower probability. This leads to more diverse responses because the model isn't always choosing the highest probability word that comes next.

Provide examples

Examples help show the model what you'll send to it and what you expect in response. Remember that you're building a prompt for a specific use case, so your examples should help support that use case. Refer to the following table for examples:

Use case	Example inputs	Example outputs
Detect contact information in an email	Email text	List of phone numbers and email addresses
Summarize complex content, like a legal brief	Legal brief text	Short, easy to read summary
Generate FAQs about a new policy	Part of the policy	Two to three questions and answers related to this part of the policy

Select a model

Model selection determines which generative AI model an AI skill uses to process prompts or documents. Each model offers different strengths. Some models prioritize speed, while others provide more advanced reasoning. Choosing the right model helps tailor your AI skill to your application's performance and accuracy needs.

To select a model when you configure an AI skill:

1. Within the AI skill, locate the Settings pane.
2. Under Provider, select the AI provider you want to use. Available options include Appian and any external providers configured by your administrator, such as Amazon Web Services or Microsoft Azure. The default provider set by your administrator is labeled with a Default tag.
3. Under Model Selection, click the dropdown to see a list of available models. The list dynamically filters based on the selected provider. When you configure a document-based AI skill, the available models change based on your document type selection: Text only or Visual elements & text.
4. Choose a model from the list. The default model when you create a new AI skill is the recommended model.

Provider and model considerations

Keep the following points in mind when selecting a provider and model:

• Provider availability: Only providers configured and enabled by your administrator appear in the provider list.
• Model availability: Only models supported by Appian and enabled by your administrator are available for selection, even if additional models are enabled in your external provider account.
• Provider changes: If an administrator disables a provider or model that your AI skill uses, the skill will fail during execution. Monitor your AI skills to ensure they continue to use active providers and models.
• AI usage: Models from external providers count toward AI usage tracking and limits within the platform, similar to Appian-provided models.
• Azure model performance: If your environment uses Microsoft Azure OpenAI, the speed and quality of an AI action within a process depends on the Azure deployment your administrator configured for each capability category. Refer to your Azure model documentation to understand the performance characteristics of each deployment before selecting a model.
• External provider support: Amazon Web Services (AWS) and Microsoft Azure are supported as external providers. See AI Services for configuration details.
• PDF page limits: Claude 4+ models (Claude 4.6 Sonnet, Claude 4.5 Sonnet, Claude 4 Sonnet, Claude 4.5 Haiku) and Amazon Nova 2 Lite support up to 100 pages for the "Visual elements & text" document processing mode. Azure OpenAI GPT models support up to 100 pages and a maximum file size of 24 MB. Legacy models support up to 20 pages. See PDF processing for details.

Enable extended thinking

Extended thinking enhances reasoning for complex tasks by prompting the model to work through its approach more thoroughly and explain the steps it takes. When enabled, the model breaks down intricate tasks into logical, sequential steps before generating a final response.

To enable extended thinking:

1. Within the AI skill, locate the Settings pane.
2. Select the Enable extended thinking toggle.
3. Click SAVE.

Extended thinking is only available for specific models. If your selected model doesn't support extended thinking, the toggle is disabled.

Test the prompt

Before you integrate the skill into a process, you want to be confident the model interprets the prompt properly and returns the results you expect. You can test the prompt using sample text to verify it works as you expected directly within the AI skill object.

To test the prompt:

1. Within the AI skill, locate the Test Prompt area.
2. Add sample text into the Test Input field.
3. Click TEST. The AI model returns a response based on your prompt and test input. The test results panel also displays the number of AI Actions consumed by the test execution.
4. (Optional) If extended thinking is enabled, select Show Thinking to view the reasoning the model used to generate the response.

The AI Actions Consumed value shown during testing reflects the same metric used during smart service execution, so you can estimate usage and related costs before deploying the skill to a process.

If the response meets your requirements, you're ready to use the skill in a process. If you want to continue experimenting, you can alter the prompt, adjust the temperature, or add more examples to help improve the model's responses.

Use the skill

Now you're ready to use your prompt in a process.

Add the Execute Generative AI Skill smart service and configure it to call your new skill. To route requests to a specific custom model at runtime, use the Runtime Model parameter, which accepts custom model identifiers.

Review the feature's compliance to ensure it aligns with your organization's security requirements.

Feedback