Agentic development and deployment

  • Release version: Australia
  • Updated June 5, 2026
  • 3 minutes to read
    • Summarize
    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 Agentic development and deployment

    Agentic development and deployment on the ServiceNow AI Platform streamlines the creation, review, testing, and deployment of applications. After generating and expanding an app using AI-driven tools, the deployment process integrates tightly into the workflow to ensure secure, efficient, and compliant delivery of updates.

    Show full answer Show less

    Deployment Workflow

    • Collaborative design: Business and IT teams collaborate on app requirements and ideas.
    • AI-driven app development: The Build Agent and ServiceNow Otto utilize process files, chat history, and diagrams to generate app updates.
    • Review and testing: Teams preview changes, revise, and conduct performance and readiness testing.
    • Developer review: A developer validates AI-generated changes before deployment.
    • Deployment approval: Deployment managers initiate approval workflows.
    • Autonomous checks: AI agents scan for issues like sensitive data exposure and remediate them automatically.
    • Final deployment: The app is deployed securely after all approvals and scans.

    Deployment Methods

    • Git-based source control integration: Supports Git workflows for version control and CI/CD, including integration with GitHub or Bitbucket. Requires Australia Patch 5 for source control in ServiceNow Studio.
    • Update sets and application packaging: Uses System Update Sets to track changes and package scoped applications for easier transport and installation, with optional use of Application Repository (AppRepo).

    Moving Apps Through Environments

    • Wrap scoped applications in update sets to move them to test instances, convert to AppRepo, and manage deployment requests via ReleaseOps or standard processes.
    • Publish apps to AppRepo using Git or update sets for distribution and pipeline integration including automated testing.
    • Register and entitle apps prior to publishing to ensure proper governance.

    Additional Deployment Tools and Capabilities

    • App Engine Management Center (AEMC): Provides governance checks validating ACLs, roles, and compliance before deployment, plus dashboards for lifecycle management.
    • ReleaseOps: Enables multi-instance deployment pipelines with customizable playbooks, automated testing (ATF), approval gates, and the ability to schedule controlled rollouts.
    • ServiceNow SDK: Allows local development and integration with off-instance CI/CD processes through CLI commands to push or install applications on ServiceNow instances.
    • Automated Test Framework (ATF): Supports generation and execution of tests to validate app functionality after changes, integrated with development tools and deployment pipelines.

    This comprehensive deployment framework empowers ServiceNow customers to efficiently develop, validate, and deploy AI-generated applications with strong governance, automation, and integration capabilities across their enterprise environments.

    After you generate and expand an app on the ServiceNow AI Platform, the deployment process is tightly integrated into the workflow.

    Workflow for deploying an agentically developed app

    After development, review, and testing are complete, a typical deployment workflow includes the following steps:
    1. Collaborative design: Business owners and IT collaborate on requirements and ideas using their preferred tools.
    2. AI-driven app development: Build Agent and ServiceNow Otto process files, chat history, and diagrams to generate and implement app updates.
    3. Review and testing: Teams preview updates, make revisions, and run rounds of performance and readiness testing.
    4. Developer review: A developer reviews the AI-generated changes, compares versions, and confirms the changes are ready for deployment.
    5. Deployment approval: The project is handed off to a deployment manager, who initiates the deployment approval process.
    6. Autonomous checks: AI agents automatically scan for issues (such as sensitive data exposure or model integrity problems) and remediate them before deployment.
    7. Final deployment: After all readiness scans and approvals, the new app is deployed securely and efficiently.

    Deployment options

    Build Agent supports the following deployment methods for apps created and edited with agentic development:
    • Git-based source control integration: ServiceNow supports Git-based workflows for version control and CI/CD.
    • Update sets and application packaging: Standard ServiceNow deployment uses System Update Sets to track changes.
      • Advanced guidance includes packing update sets into scoped applications for easier transport and installation across instances, for example using Application Repository (AppRepo).
      • For more information on System Update Sets, see System update sets.

    Workflows for moving apps through higher instances

    After you create an app using Build Agent, you have several options to move the app to the test instance.
    1. Wrap the entire scoped application in an update set. The workflow is as follows:
      1. Go to the Custom Applications list, select an app and swap to its scope.
      2. Convert the app to AppRepo.
      3. Publish the update set with demo data.
      4. Put the update set in a deployment request for ReleaseOps, or follow your standard update set process for deployment.
    2. Publish the app to AppRepo:
      • You can use a Git-based process or update sets to publish to AppRepo.
      • Scoped apps, as well as apps that are ready for testing, can be published to the AppRepo for distribution across environments.
      • After an app is in AppRepo, you can move it through a ReleaseOps pipeline. If ATF tests are included in the pipeline, they automatically run.
      • Register and entitle apps before publishing.
      • For more information on Application Repository, see ServiceNow application repository.

    Additional deployment tools and capabilities

    The ServiceNow AI Platform has additional deployment tools that include the following tools:
    • App Engine Management Center (AEMC):
      • After developing an app, submit it to AEMC for governance checks.
      • AEMC validates ACLs, roles, and compliance settings before deployment.
      • Use ReleaseOps pipelines to move apps through environments with ATF tests and approval gates.
      • AEMC provides dashboards for monitoring deployments and managing app versions throughout the lifecycle.
      • For more information on AEMC, see Using the App Engine Management Center.
    • ReleaseOps:
      • Move changes from development to production through multiple instances using customizable playbooks.
      • Automate preview, commit, and validation of update sets before deployment.
      • Run Automated Test Framework (ATF) tests as part of the pipeline to validate quality.
      • Deploy changes immediately or schedule releases for controlled rollouts.
      • Enforce checks, scans, and approvals before production deployment.
      • For more information on ReleaseOps, see ReleaseOps.
    • ServiceNow SDK:
      • Use the ServiceNow SDK to move applications to and from your instance to your local machine. You can integrate the ServiceNow SDK with your off-instance CI/CD process if you have one.
      • Install the ServiceNow SDK locally and use the command line interface (CLI).
      • Authenticate to a ServiceNow instance from the ServiceNow SDK.
      • Push to or install an application on the authenticated instance from your local environment.
    • Automated Test Framework (ATF)
      • Tests can be generated by Build Agent and executed in ServiceNow Studio or ServiceNow IDE to confirm functionality after changes.
      • For more information on ATF, see Automated Test Framework (ATF).