Skip to content
NexusBook a demo

Enterprise use case · GitHub + ServiceNow + OpenAI

Change summary for approvers

Approvers must connect code changes, test evidence and change-ticket descriptions before they can assess what is actually being requested. Change authorities receive a concise, source-linked account of the proposed change, its stated impact and its unresolved risks.

Systems involved

What each system contributes

GitHub

2 actions

  • Repository search1
  • Check errors1

Trigger source: GitHub

ServiceNow

55 actions

  • Change Management10

Trigger source: ServiceNow

OpenAI

25 actions

  • Chat Completions8

GitHub pre-built actions cover repository search and check errors. Revision and diff retrieval uses an authorized GitHub API or Rules + Code step in the same workflow. The model step runs against the AI provider you choose, including Anthropic, Google Gemini or a model you host yourself. Anything outside a pre-built action runs through an API, protocol or database call, or as Rules + Code in Python, Bash, Go, Java, .NET, PHP, PowerShell or JavaScript/TypeScript, with the same approvals and audit.

Architecture

One governed run

Trigger: ServiceNow changeIntegration: GitHubAI step: model of your choiceRules + CodeHuman ApprovalAudit

How it starts: A ServiceNow change owner releases a proposed software change for approval review.

Step by step

How the process runs

  1. 01

    Gather the exact revision

    Read the change record and the authorized GitHub diff, repository and check context for the submitted revision.

  2. 02

    Summarize the change

    The selected model describes affected components, interfaces and configuration using source references rather than relying on the author's summary alone.

  3. 03

    Identify evidence gaps

    The model distinguishes observed changes from claimed effects and calls out absent tests, missing dependencies or undocumented assumptions.

  4. 04

    Check the review package

    Rules verify revision consistency and mandatory evidence before attaching the proposed brief to ServiceNow.

  5. 05

    Record the approval decision

    The named change authority reviews the sources and brief, then approves, rejects or requests more information through the designated process.

Control

Human decision and exception path

Human decision

The summary does not authorize deployment, access changes or changes to production data. The named change authority approves the specific revision before any separately configured implementation step proceeds.

Exception path

  • A new commit invalidates the earlier revision summary and returns the package for review.
  • Missing diffs or checks are shown as gaps rather than reported as successful tests.
  • Unsupported impact claims remain unverified until the technical owner supplies evidence.

Evidence

What gets audited

  • GitHub revision and ServiceNow change reference.
  • Retrieved diff and check references under the permitted logging scope.
  • Model summary, evidence gaps and review-package checks.
  • Approver decision, authorized revision and timestamp.

Nexus runs on premises, in private or sovereign cloud, or air-gapped. Credentials stay in the vault, roles decide who can approve, and the execution record stays inside your deployment boundary.

Sovereignty and deployment

Related

Patterns, connectors and next steps

FAQ

Questions about this use case

Does Nexus replace any of these systems?
No. Each system keeps its own records. Nexus runs the process between them and records each decision.
How does the run start?
A ServiceNow change owner releases a proposed software change for approval review.
What if a step is not covered by a pre-built action?
GitHub pre-built actions cover repository search and check errors. Revision and diff retrieval uses an authorized GitHub API or Rules + Code step in the same workflow. The model step runs against the AI provider you choose, including Anthropic, Google Gemini or a model you host yourself. Anything outside a pre-built action runs through an API, protocol or database call, or as Rules + Code in Python, Bash, Go, Java, .NET, PHP, PowerShell or JavaScript/TypeScript, with the same approvals and audit.
Does a readable summary prove that a change is safe?
No; it helps the approver inspect the evidence but does not replace testing, technical review or the change decision.

Use case review

Discuss this use case

Bring your version of the process. We will map the systems, approvals and evidence it needs.