Enterprise use case · ServiceNow + Amazon Web Services + OpenAI + Jira
Proposed remediation for an incident
Operators assemble possible fixes from incident notes and resource context without a consistent record of why one action was chosen. Incident owners receive a source-backed remediation plan ready for technical review and the required change approval.
Systems involved
What each system contributes
61 actions
- EC215
- Lambda10
- IAM21
Trigger source: AWS S3, AWS SNS, AWS SQS, AWS CloudWatch, AWS CloudWatch Events
25 actions
- Chat Completions8
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
How it starts: An assigned ServiceNow incident is released for remediation planning.
Step by step
How the process runs
- 01
Establish the affected scope
Read the permitted incident context, linked Jira work and relevant AWS resource state.
- 02
Retrieve approved procedures
Select the applicable runbook or previously reviewed resolution from the authorized incident or work-item context.
- 03
Propose the remedy
The selected model drafts a scoped plan with prerequisites, expected checks, unresolved assumptions and a proposed rollback path.
- 04
Validate the actions
Rules limit the plan to permitted resources and operations and flag commands or steps that need additional technical review.
- 05
Execute after change approval
A separate authorized AWS action or controlled code step performs only the approved operation and returns its result to the incident record.
- 06
Check the result
Deterministic health checks and operator review determine whether to continue, escalate or request the approved rollback.
Control
Human decision and exception path
Human decision
The named change authority approves the exact action scope before any infrastructure or access change. A changed resource state or revised plan can require revalidation and renewed approval.
Exception path
- Unsupported actions or an unverified rollback proposal stop the plan at technical review.
- Resource drift invalidates affected assumptions before execution proceeds.
- Failed checks stop subsequent actions and invoke only the rollback or escalation path already authorized by policy.
Evidence
What gets audited
- Incident, Jira and AWS context references.
- Model proposal, cited procedure and validation results.
- Change approval, authorized scope and decision time.
- Executed requests, responses and post-change checks.
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 deploymentRelated
Patterns, connectors and next steps
- AWS-ServiceNow integration
Integration pattern · Cloud events become enriched incidents and approved changes.
- ServiceNow connector
55 pre-built actions
- Amazon Web Services connector
61 pre-built actions
- OpenAI connector
25 pre-built actions
- Jira connector
36 pre-built actions
- Infrastructure orchestration
Operations runbooks with approval and evidence.
- Order to cash across CRM, ERP and billing
Another enterprise use case.
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?
- An assigned ServiceNow incident is released for remediation planning.
- What if a step is not covered by a pre-built action?
- 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 the model run its own suggested commands?
- No; the proposal is reviewed, and a distinct workflow step executes the authorized operation with scoped credentials.
- Can the model create a rollback guarantee?
- No; rollback feasibility must be verified for the action, and unsupported recovery assumptions remain blockers for review.
Use case review
Discuss this use case
Bring your version of the process. We will map the systems, approvals and evidence it needs.