AI agents and agentic workflows
Design, deploy, and evaluate AI agents and agentic workflows that autonomously complete multi-step tasks on the ServiceNow AI Platform.
Agentic AI
AI agents and agentic workflows pursue a goal across multiple steps with limited human intervention, instead of completing one task at a time. To learn what they are, why they matter, and when to use them rather than a generative AI skill, see About AI agents and agentic workflows.
AI Agents library
Use the ServiceNow AI agents library to review all ServiceNow® created AI agents. Use this reference to understand each agent's purpose, workflow, and configuration.
AI Agent Studio
Use the AI Agent Studio environment to create, manage, and test AI agents and agentic workflows. Putting AI agents to work follows four steps: create an agentic workflow that defines the problem and situation your agents work on, create AI agents equipped with the tools to execute the necessary tasks, test the workflow and make adjustments until it performs as expected, and then deploy your agents across your organization.
AI Agent Studio also provides ready-made agentic workflows and AI agents you can use as-is or incorporate into custom workflows, an analytics dashboard for reviewing agent usage and performance, and testing tools that range from single manual executions to automated agentic evaluations. From the Settings page, you can enable AI Guardian to apply offensiveness detection and prompt injection protection to your agents.
To create, test, and deploy AI agents and agentic workflows in one environment, with ready-made options, analytics, and built-in testing, see AI Agent Studio (legacy).
Now Assist agentic workflows
The ServiceNow AI Platform provides prebuilt platform agentic workflows and AI agents that address common scenarios, such as approval assistance, issue readiness, and request status. Additional AI agents are provided with some Integration Hub spokes and plugins, preconfigured and callable from interfaces such as AI Agent Studio and Virtual Agent. These prebuilt workflows and agents give you a faster path to value than building from scratch, and you can incorporate them as-is or customize them to match your processes.
To put prebuilt agentic workflows and AI agents to work for common scenarios such as approvals, issue readiness, and request status, see Agentic workflows.
AI Desktop Actions
AI Desktop Actions extends AI agent capabilities to applications outside the ServiceNow AI Platform. In the no-code Design workspace, you create desktop actions by auto-capturing or manually recording a series of steps you perform on your computer, such as clicking buttons, entering text, or selecting from drop-downs across different applications. You then save that sequence as a reusable desktop action. After you activate a desktop action, it becomes available as a tool in AI Agent Studio, enabling your AI agents to interact with desktop applications as part of an agentic workflow.
To record reusable desktop actions that let your AI agents work in applications outside the ServiceNow AI Platform, see .
--- title: AI agents and agentic workflows description: Design, deploy, and evaluate AI agents and agentic workflows that autonomously complete multi-step tasks on the ServiceNow AI Platform. locale: en-US canonical_url: https://www.servicenow.com/docs/r/intelligent-experiences/platai-ai-agents-agentic-workflows.html release: brazil topic_type: concept last_updated: "2026-09-10" reading_time_minutes: 3 keywords: [AI agents, agentic workflows, agentic AI, AI Agent Studio, platform AI agents, Agentic Desktop] breadcrumb: [Enable AI Experiences] --- # AI agents and agentic workflows Design, deploy, and evaluate AI agents and agentic workflows that autonomously complete multi-step tasks on the ServiceNow AI Platform. ## Agentic AI AI agents and agentic workflows pursue a goal across multiple steps with limited human intervention, instead of completing one task at a time. To learn what they are, why they matter, and when to use them rather than a generative AI skill, see [About AI agents and agentic workflows](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/intelligent-experiences/platai-understanding-ai-agents-agentic-workflows.md). ## AI Agents library Use the [ServiceNow AI agents library](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/intelligent-experiences/ai-agent-landing-page.md) to review all ServiceNow® created AI agents. Use this reference to understand each agent's purpose, workflow, and configuration. ## AI Agent Studio Use the AI Agent Studio environment to create, manage, and test AI agents and agentic workflows. Putting AI agents to work follows four steps: create an agentic workflow that defines the problem and situation your agents work on, create AI agents equipped with the tools to execute the necessary tasks, test the workflow and make adjustments until it performs as expected, and then deploy your agents across your organization. AI Agent Studio also provides ready-made agentic workflows and AI agents you can use as-is or incorporate into custom workflows, an analytics dashboard for reviewing agent usage and performance, and testing tools that range from single manual executions to automated agentic evaluations. From the Settings page, you can enable AI Guardian to apply offensiveness detection and prompt injection protection to your agents. To create, test, and deploy AI agents and agentic workflows in one environment, with ready-made options, analytics, and built-in testing, see [AI Agent Studio \(legacy\)](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/intelligent-experiences/na-ai-agents.md). ## Now Assist agentic workflows The ServiceNow AI Platform provides prebuilt platform agentic workflows and AI agents that address common scenarios, such as approval assistance, issue readiness, and request status. Additional AI agents are provided with some Integration Hub spokes and plugins, preconfigured and callable from interfaces such as AI Agent Studio and Virtual Agent. These prebuilt workflows and agents give you a faster path to value than building from scratch, and you can incorporate them as-is or customize them to match your processes. To put prebuilt agentic workflows and AI agents to work for common scenarios such as approvals, issue readiness, and request status, see [Agentic workflows](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/intelligent-experiences/sn-aia-use-cases-list.md). ## AI Desktop Actions AI Desktop Actions extends AI agent capabilities to applications outside the ServiceNow AI Platform. In the no-code Design workspace, you create desktop actions by auto-capturing or manually recording a series of steps you perform on your computer, such as clicking buttons, entering text, or selecting from drop-downs across different applications. You then save that sequence as a reusable desktop action. After you activate a desktop action, it becomes available as a tool in AI Agent Studio, enabling your AI agents to interact with desktop applications as part of an agentic workflow. To record reusable desktop actions that let your AI agents work in applications outside the ServiceNow AI Platform, see .
| title | AI agents and agentic workflows |
| description | Design, deploy, and evaluate AI agents and agentic workflows that autonomously complete multi-step tasks on the ServiceNow AI Platform. |
| locale | en-US |
| canonical_url | https://www.servicenow.com/docs/r/intelligent-experiences/platai-ai-agents-agentic-workflows.html |
| release | brazil |
| topic_type | concept |
| last_updated | 2026-09-10 |
| reading_time_minutes | 3 |
| keywords | AI agentsagentic workflowsagentic AIAI Agent Studioplatform AI agentsAgentic Desktop |
| breadcrumb | Enable AI Experiences |