--- sourceDocument: Australia Build or modify applications sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/application-development Release : - australia ft:locale : - en-US ft:publication_title : - Australia Build or modify applications ft:clusterId : - cadev bundleId : - cadev workflow : - Development, Data, and Analytics --- # AI assets # AI assets {#ariaid-title1} Release version: Australia Updated March 31, 2026 ![](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 AI assets The ServiceNow AI Platform enables you to create three types of AI assets---skills, AI agents, and agentic workflows---each designed to add generative AI capabilities to custom applications. Understanding these asset types helps you select the right AI approach tailored to your specific use case. Show full answer Show less ## Key Features * **Skills:** Focused generative AI functions that complete a single task, such as summarizing or generating text based on input records. Skills are easily invoked via the Now Assist panel, UI buttons, or Virtual Agent, making them ideal for adding straightforward AI capabilities. * **AI Agents:** More advanced generative AI that follows instructions and uses tools to perform multiple actions and make decisions. AI agents suit scenarios requiring varied operations and adaptable solutions beyond single tasks. * **Agentic Workflows:** Orchestrated workflows that coordinate multiple AI agents working in sequence or parallel to achieve complex, end-to-end objectives. These workflows are suited for comprehensive processes like classifying requests, retrieving data, and generating responses. ## Practical Use and Customization You can start with preconfigured Now Assist Platform AI assets and customize them to fit your application's tables, fields, and workflows. Preconfigured assets are optimized for platform use cases but often need significant adaptation for custom apps. Alternatively, you can build custom AI assets from scratch tailored precisely to your application requirements. Choosing the right AI asset type enables efficient integration of generative AI into your ServiceNow custom applications, enhancing automation and user assistance capabilities. Learn about the AI assets that you can create on the ServiceNow AI Platform. There are three types of AI assets that you can create on the ServiceNow AI Platform: skills, AI agents, and agentic workflows. Each AI asset type serves a different purpose and suits different use cases. Learning about what each asset is can help you decide how to approach adding AI to your custom applications. For detailed guidance about choosing the right asset type for your use case, see [Choosing the right AI capability](https://servicenow-prod.fluidtopics.net/07tfwApdK4YwvH2QectFew "Choosing the right AI capability for your custom application is important. Learn about the differences between skills, AI agents, and agentic workflows, use cases for each, and when to choose one capability over another."). Skills : A skill is a generative AI capability that completes a specific task for a defined use case. You provide the skill with an input, typically a record from a table in your application, and the skill uses an underlying large language model (LLM) to generate an output, such as a summary or a block of text. Skills are the most contained AI asset type and are a good starting point if you want to add a single, focused AI capability to your custom app. You can invoke skills conversationally through the Now Assist panel, through UI actions such as buttons on a record, or through Virtual Agent. AI agents : An AI agent is a generative AI capability that follows a set of LLM instructions and uses tools to carry out a role. Unlike skills, which complete a single task, AI agents can make decisions and take a series of actions based on their instructions and the data available to them. AI agents are well suited to use cases where there might be more varied situations with different operations and solutions. Agentic workflows : An agentic workflow combines one or more AI agents to accomplish a broader objective. An AI Agent Orchestrator coordinates the work between agents, directing each one to perform its part of the overall task. Agentic workflows are appropriate when the outcome requires multiple agents working in sequence or in parallel. For example, classifying a request, retrieving relevant records, and generating a recommended response in a single end-to-end workflow. For each asset type, you can use preconfigured Now Assist Platform assets as a starting point and modify their underlying structure to fit your app's use case. Or you can build custom AI assets tailored to your app from scratch. Preconfigured assets are designed for existing platform use cases and typically require significant reconfiguration to work within custom applications. Custom AI assets can be designed for your specific tables, fields, and workflows from the beginning. For more information, see [AI capabilities for enhancing custom applications](https://servicenow-prod.fluidtopics.net/HzIGfd44bRHiStMW9JwmpA "Learn about the AI capabilities available with ServiceNow Otto for App Engine that you can use to enhance custom applications.").