Understanding generative AI skills
Generative AI skills use large language models to process your enterprise data to complete a single, well-defined task within a workflow, such as summarizing a record, drafting a reply, or recommending a next action. Skills reduce manual effort in work your users already do, without changing how they work.
A generative AI skill completes one task that is triggered in the moment that you need it. The skill sends a prompt and your data to a large language model, then returns a result you can use directly in the record or conversation you're working in, such as a summary, a drafted message, or a recommended action. Because each skill targets a specific task, you can adopt skills one at a time and see results without redesigning your processes.
Why generative AI skills matter
Skills reduce the manual, repetitive effort that slows down everyday work. A user who summarizes a long case in seconds, or who drafts resolution notes from a single prompt, spends less time writing and searching and more time resolving the issue. Skills run inside the interfaces your users already work in, so the productivity gain comes with little training and little change to established habits.
Skills also give your organization a measured entry point into AI. You can start with a common, self-contained task, review the result, and add more skills as your users build confidence.
Generative AI skills compared with AI agents
A skill and an AI agent solve different kinds of problems. A skill performs one bounded task on demand and returns control to the user, who decides what to do with the result. An AI agent, working within an agentic workflow, pursues a broader goal across multiple steps and can plan, act, and adapt with limited human intervention.
Consider a service request. A skill can summarize the request or draft a reply when a user asks for it. An agentic workflow can take the request further on its own by interpreting it, gathering the information it requires, completing the required actions, and reporting the outcome. Skills are the building blocks for individual tasks, and agents coordinate those capabilities to carry a process from start to finish.
When to use generative AI skills
Choose a generative AI skill when the work is a single, predictable task that a person starts and reviews, such as summarizing, drafting, translating, or recommending content. Skills fit well when you want an early, measurable result, when the user stays in control of each result, and when you're introducing generative AI to a team for the first time.
When an outcome requires several coordinated steps, decisions that depend on changing conditions, or actions across multiple systems without a person driving each one, an AI agent and an agentic workflow are the better fit.
--- title: Understanding generative AI skills description: Generative AI skills use large language models to process your enterprise data to complete a single, well-defined task within a workflow, such as summarizing a record, drafting a reply, or recommending a next action. Skills reduce manual effort in work your users already do, without changing how they work. locale: en-US canonical_url: https://www.servicenow.com/docs/r/intelligent-experiences/platai-understanding-generative-ai-skills.html release: brazil topic_type: concept last_updated: "2026-09-10" reading_time_minutes: 2 keywords: [generative AI skills, Now Assist skills, generative AI, skills versus AI agents, large language model, when to use generative AI skills] breadcrumb: [Generative AI skills, Enable AI Experiences] --- # Understanding generative AI skills Generative AI skills use large language models to process your enterprise data to complete a single, well-defined task within a workflow, such as summarizing a record, drafting a reply, or recommending a next action. Skills reduce manual effort in work your users already do, without changing how they work. A generative AI skill completes one task that is triggered in the moment that you need it. The skill sends a prompt and your data to a large language model, then returns a result you can use directly in the record or conversation you're working in, such as a summary, a drafted message, or a recommended action. Because each skill targets a specific task, you can adopt skills one at a time and see results without redesigning your processes. ## Why generative AI skills matter Skills reduce the manual, repetitive effort that slows down everyday work. A user who summarizes a long case in seconds, or who drafts resolution notes from a single prompt, spends less time writing and searching and more time resolving the issue. Skills run inside the interfaces your users already work in, so the productivity gain comes with little training and little change to established habits. Skills also give your organization a measured entry point into AI. You can start with a common, self-contained task, review the result, and add more skills as your users build confidence. ## Generative AI skills compared with AI agents A skill and an AI agent solve different kinds of problems. A skill performs one bounded task on demand and returns control to the user, who decides what to do with the result. An AI agent, working within an agentic workflow, pursues a broader goal across multiple steps and can plan, act, and adapt with limited human intervention. Consider a service request. A skill can summarize the request or draft a reply when a user asks for it. An agentic workflow can take the request further on its own by interpreting it, gathering the information it requires, completing the required actions, and reporting the outcome. Skills are the building blocks for individual tasks, and agents coordinate those capabilities to carry a process from start to finish. ## When to use generative AI skills Choose a generative AI skill when the work is a single, predictable task that a person starts and reviews, such as summarizing, drafting, translating, or recommending content. Skills fit well when you want an early, measurable result, when the user stays in control of each result, and when you're introducing generative AI to a team for the first time. When an outcome requires several coordinated steps, decisions that depend on changing conditions, or actions across multiple systems without a person driving each one, an AI agent and an agentic workflow are the better fit.
| title | Understanding generative AI skills |
| description | Generative AI skills use large language models to process your enterprise data to complete a single, well-defined task within a workflow, such as summarizing a record, drafting a reply, or recommending a next action. Skills reduce manual effort in work your users already do, without changing how they work. |
| locale | en-US |
| canonical_url | https://www.servicenow.com/docs/r/intelligent-experiences/platai-understanding-generative-ai-skills.html |
| release | brazil |
| topic_type | concept |
| last_updated | 2026-09-10 |
| reading_time_minutes | 2 |
| keywords | generative AI skillsNow Assist skillsgenerative AIskills versus AI agentslarge language modelwhen to use generative AI skills |
| breadcrumb | Generative AI skillsEnable AI Experiences |