AI Skill Object

Overview

An AI skill is a design object that enables you to build and configure artificial intelligence (AI) capabilities using Appian.

This page discusses the basic features and functionality of the AI skill design object.

Create an AI skill

To create an AI skill:

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

   

2. Select the skill type you want to create.

   

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.
7. Click CREATE FIRST MODEL.

Skill types

As you create an AI skill, you'll select the type of skill that suits your purpose.

General purpose

If you don't see an AI skill suited to your use case, you can always use the Prompt Builder. Use this skill 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. Use this skill to summarize long or complex text, generate new content, and more!

Classification

• Text: Identify text based on certain traits. You'll tell us about the types of text you want to classify in your business process and provide samples of each type.
• Emails: Identify an email's type based on certain traits. You'll tell us about the types of emails you want to classify in your business process and provide samples of each type. For example, this skill can help you classify and route inquiries, feedback, and help requests that you receive through one inbox.
• Documents: Identify a document's type based on certain traits. You'll tell us about the types of documents you want to classify in your business process and provide samples of each type. For example, this skill can help you classify and route invoices, purchase orders, and receipts that you receive through one channel.

Extraction

• Text or Emails: Extract data from text or emails. You'll tell us what you want to extract and how you want that information to appear in the output.
• Documents: Extract data from a document.
  • Semi or Highly structured: You'll tell us about the structure of the type of document you want to extract data from, such as invoices. For example, this skill can help you extract data from fields commonly found in certain documents. For invoices, this might be the invoice number, due date, and total.
  • Unstructured: You'll tell us what you want to extract from an unstructured document and how you want that information to appear in the output. For example, this skill can help you extract data from large or free-flowing paragraphs of text, like contracts.
  • Advanced IDP Tools: Tell us the type of document data you want to process, and we'll take care of the rest. The output is structured data that developers can use to add additional processing or validation if needed. For example, you could use expression rules or a Prompt Builder AI skill to map the outputs to fields in a record type.

Generation

• Text: Create text for any purpose. You'll tell us what text you want to generate and whether you want the output to include any specific logic, format, or other requirements.

Summarization

• Text, Emails, or Documents: Summarize text from any source, including emails and documents. You'll tell us what you want to summarize and how you want the summary to appear.

PII Detection

• Text, Emails, or Documents: Extract personally identifiable information (PII) in a document, email, or text block. You'll tell us the personally identifiable information (PII) you want to extract, such as names or phone numbers.

You'll configure your skill based on its type.

Design guidance

The AI skill design object is not versioned. However, as you train your skill, you can review the results and create a new model using additional sample data. During this refinement process, each model is preserved in the training summary.

You can access the training summary to view details on model creation history and how the skill was refined to its current state.

Using an AI skill in a process

After you train the skill and determine a model's predictions are suitable for your purposes, you'll publish the model. The published model is what the corresponding smart service uses when it calls the AI skill in a process model.

You'll call an AI skill in a process model using the smart services in the AI Skill menu of the palette:

• Classify Documents
• Classify Emails
• Extract from Document
• Execute Generative AI Skill

Monitor AI skill execution

AI skills do not have a dedicated Monitor tab. To track AI skill execution time, consumption, and model usage, use a combination of process monitoring, smart service outputs, and environment logs.

View execution time

AI skill execution time is visible through the process model that calls the skill. The Execute Generative AI Skill smart service runs as a node in a process model, and its execution duration is captured by the platform's standard process monitoring tools.

To view AI skill execution time:

• Process Activity tab: In the Monitor view, locate the process instance that contains the AI skill node. The Start Time and End Time columns show the overall process duration. Click the process name to open it in the Process Modeler, where you can inspect individual node details.
• Autoscaled Process Activity tab: If the process model has autoscale enabled, use the Autoscaled Process Activity tab to monitor how AI skill nodes perform at scale.
• Process variables: To capture per-node execution time programmatically, use a script task to record a timestamp before and after the Execute Generative AI Skill node. Store the difference in a process variable for use in reports or dashboards.

Track AI Actions consumption

Each AI skill execution consumes AI Actions, which are the unit of measurement for AI consumption in Appian.

AI Actions consumption data is available at the following levels:

• Per-execution (smart service output): The Execute Generative AI Skill smart service returns an AI Actions Consumed output for each execution. Store this value in a process variable to build dashboards or track costs over time.
• Per-execution (test panel): The AI Skill test panel displays AI Actions Consumed after each test, so you can estimate costs during development before deploying the skill to a process.
• Per-execution (AI input/output log): The AI input/output log records AI Actions consumed for each execution, along with the optional usage group, when enabled by an administrator.
• Per-group attribution: Configure the Usage Groups input on the smart service to attribute AI Action consumption to specific Appian groups for internal chargeback and cost allocation.
• Environment-wide (MyAppian): The Adoption page in MyAppian displays aggregate AI Action consumption across environments and AI features.

Review AI usage in environment logs

Administrators can review AI skill activity using the environment's AI logs. These logs track all AI feature executions, including AI skills.

• AI usage log (audit/artificial_intelligence.csv): Records when AI features are invoked, including the execution timestamp, user UUID, object UUID, and feature name. This log is always active and does not capture input or output content. See AI usage logs for the full log structure.
• AI input/output log (perflogs/ai/ai_input_output_activity.log): Records the inputs sent to the AI model and the model's response, along with the model ID and a correlation ID for tracing related events. Administrators must enable this log in the Administration Console. See AI input/output logs for details.

Identify the model used for an execution

AI skills use Appian-managed model identifiers—not API keys or tokens that you retrieve as credentials. Appian does not expose model access tokens or authentication secrets to developers.

To determine which model an AI skill used for a specific execution:

• Smart service output: The Execute Generative AI Skill smart service returns a Model output that identifies the specific generative AI model used for each execution.
• AI input/output log: The AI input/output log records the model ID for generative AI executions. For AI skills using a custom provider, the model ID appears as custom/<model_uuid>.

AI skill throughput and scaling

Appian manages AI skill throughput at the platform level. There is no configurable setting for the maximum number of concurrent AI skill executions.

The following constraints apply to AI skill execution:

• Standard mode timeout: The AI model must respond within 4.5 minutes.
• Long Running mode timeout: The AI model must respond within 60 minutes. Use this mode for complex reasoning, large outputs, or high-volume processing.
• Autoscaled process timeout: In autoscaled processes, the timeout is 1.5 minutes regardless of execution mode.
• Rate limiting: If the AI model receives too many simultaneous requests, executions may fail with the error "The AI model is currently processing too many other requests. Try again later." Reduce the frequency of concurrent calls if you encounter this error.

To scale AI skill execution in high-volume scenarios:

• Use Long Running mode with autoscale: For processes that run many AI skill executions in parallel, combine Long Running mode with autoscale to avoid standard mode timeouts and handle concurrent executions.
• Distribute load over time: Stagger process starts or use timers to avoid sending many AI skill requests simultaneously.
• Monitor autoscaled processes: If your process model uses autoscale, monitor performance on the Autoscaled Process Activity tab and test thoroughly before deploying to production.
• Contact Appian Support: If rate limits or timeouts are blocking your use case, contact Appian Support to discuss your throughput requirements.

Delete

Deleting an AI skill prevents users from further viewing or editing it or the models it contains. Additionally, any smart service configured to use the AI skill will no longer work as intended.

AI skills can be deleted by users with Administrator permissions to it.

To delete an AI skill:

1. In your application, go to the Build view.
2. Select the checkbox of the AI skill you want to delete.
3. Click Delete.
4. Review any object dependents and click DELETE.

Security

Actions	Administrator	Editor	Viewer	Deny
Execute the skill	Yes	Yes	Yes	No
View the skill	Yes	Yes	Yes	No
Update the skill	Yes	Yes	No	No
Save the skill	Yes	Yes	No	No
View the security	Yes	Yes	No	No
Create a model in the skill	Yes	Yes	No	No
Publish a model in the skill	Yes	Yes	No	No
Configure the smart service to use the skill	Yes	Yes	No	No
Update the security	Yes	No	No	No
Delete the skill	Yes	No	No	No

Deployment behavior

You can deploy the AI skill alongside other design objects in your Appian application. This section describes what is deployed with the object and what isn't.

AI Skill	Deployed	Not Deployed
Document extraction AI skill	Document structure Published model Training metrics for published model Learned mappings from reconciliation tasks	Documents used to train the model Unpublished models
Document classification AI skill	Document types Published model Training metrics for published model	Documents used to train the model Unpublished models
Email classification AI skill	Email types Published model Training metrics for published model	Emails used to train the model Unpublished models
Generative AI skills	Prompt Temperature Example inputs and outputs	N/A

Compliance

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

Feedback