---
sourceDocument: Brazil Build or modify applications
sourceDocumentLink: https://www.servicenow.com/docs/r/application-development

 Release :

    - brazil

ft:locale :

    - en-US

ft:publication_title :

    - Brazil Build or modify applications

ft:clusterId :

    - cadev

bundleId :

    - cadev

workflow :

    - Development, Data, and Analytics


---

# Workflow

# Build Agent workflow {#ariaid-title1}

Release version: Brazil  
Updated September 22, 2026  
![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 minute to read  
The Build Agent workflow automates building applications, testing, and deploying update sets on the ServiceNow AI Platform. Build Agent streamlines development by handling code compilation, quality checks, and deployment steps without manual intervention.
Note:  
For large or complex implementations, you can use Autonomous Engineer instead of the conversational workflow that follows. Autonomous Engineer accepts your requirements up front, interviews you to resolve ambiguity, generates a plan with work items, and builds all work items in parallel in the background. For more information, see [Autonomous Engineer in Build Agent](https://www.servicenow.com/docs/M5MXvfne6ropY8H9RXIRZQ "Autonomous Engineer is an agentic worker that augments product development capacity across the SDLC. It intakes requirements, applies platform best practices through agent packs, and generates an implementation plan before building work items in parallel through Build Agent.").  
A general workflow for using Build Agent in either ServiceNow Studio or the ServiceNow IDE is the following:

1. Make sure that everything you need is properly configured in the settings, such as supported MCP server connections. For more information, see [Build Agent configuration](https://www.servicenow.com/docs/vGosc~AAxXHesTaIyucaTA "Build Agent connects your instance to AI-powered design workflows. Set up the required application and MCP server connections to start using it.").
2. Open ServiceNow Studio. Use the central chat area on the home page to start a new conversation, or select the Conversations icon ![]() in the Navigator panel to open an existing conversation.
3. Describe what to create or change in natural language.
4. Let Build Agent parse requirements and propose the application and files to create or modify.
5. Build Agent edits code or metadata or scaffolds a new application.
6. Review proposed edits, diffs, and summaries, and approve or adjust before applying changes. Review checkpoints and manual edit update sets. In ServiceNow Studio, view generated app details from the Apps tab, and inspect the source code from the Explorer tab.
7. Iterate until the desired metadata changes are complete. For more information, see [Supported metadata in Build Agent](https://www.servicenow.com/docs/SNALyNXdpWPwsTV6h1cDFQ "Metadata and app file types that Build Agent can create and manage.").
8. Prompt Build Agent to create and run Automated Test Framework (ATF) tests to verify that the tests execute as expected. Depending on your configuration, Build Agent may ask you if you want to run ATF tests. If there are failures, auto troubleshooting triages the tests and produces a regression test suite that you can use to monitor app health.
9. Instruct Build Agent to build the application; verify results in the File Navigator or Metadata Explorer.
10. Deploy the application. If you're using source control, you can push to Git. For more information, see [Deploying what you built with Build Agent](https://www.servicenow.com/docs/rb~KixtWZ4aRbYjEU6knkA "Learn about deployment methods and workflows for moving applications created with Build Agent from development to production environments. Choose the right deployment approach based on your application complexity and organizational requirements.").
{#build-agent-workflow__ol_srh_55h_b3c}

When your app is connected to Git-based source control, you can also manage source control operations from the Build Agent chat panel without switching to the VS Code UI. For more information, see .