/
/
Turn a ServiceNow alert into the right incident, once
For operations and NOC teams who want monitoring alerts raised as correctly prioritized ServiceNow incidents, by asking their AI assistant.
On this page
Connector
Connect ServiceNow, pick your AI tool and paste the prompt.
The problem
Alerts that become duplicate incidents, or none at all
Turning a monitoring alert into an incident by hand means working out impact and urgency, finding the support group and linking the alert back. It is easy to raise a second incident for something already being worked, or leave the alert firing.
Without PopdockAI
- Guess impact and urgency from the alert
- Look up the CI's support group
- Miss an open incident on the same CI
- Leave the alert unacknowledged
With PopdockAI
- Alerts that already have an incident stop the job
- Open incidents on the same CI are offered for linking
- Severity maps to impact and urgency by fixed rules
- The incident is created after you confirm, and the alert linked and acknowledged
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Raise an incident for alert `{alert-number}`, assign it to `{assignment-group}` (default: the configuration item's support group), and link and acknowledge the alert.
**Step 1: Read the alert**
Call `servicenow-get-alert` with number `{alert-number}`. Keep sys-id, severity, state, configuration-item, node, source, metric-name, description and task.
* If task is already set, the alert already has an incident: report that number and stop.
* If state is Closed, ask the user whether to continue before doing anything else.
**Step 2: Resolve the configuration item and owner**
Call `servicenow-get-configuration-item` with name set to the alert's configuration-item. Keep its support-group and environment. If `{assignment-group}` is blank use that support-group; if both are blank, ask the user.
**Step 3: Avoid a duplicate**
Call `servicenow-get-incident` with short-description.contains set to the configuration item name and opened.from yesterday, in states New and In Progress. If an open incident already covers this CI and symptom, offer to link the alert to it instead of creating a new one.
**Step 4: Map severity to impact and urgency**
* Critical: impact 1 - High, urgency 1 - High
* Major: impact 2 - Medium, urgency 1 - High
* Minor: impact 2 - Medium, urgency 2 - Medium
* Warning or lower: impact 3 - Low, urgency 3 - Low
* Raise impact one level for a Production environment CI that is not already High.
**Step 5: Confirm**
Show the proposed incident: short-description "<severity> alert on <CI>: <metric-name or first line of description>", description (alert description, node, source, alert number), configuration-item, assignment-group, impact, urgency, contact-type Monitoring if that value is listed. Ask for an explicit yes.
**Step 6: Write, in this order**
1. Call `servicenow-add-incident` once with the confirmed values.
2. Call `servicenow-get-incident` with short-description.contains set to the alert number or the CI name and opened.from today to get the new incident's number and sys-id.
3. Call `servicenow-update-alert` once with the alert sys-id, incident set to the new incident's sys-id, acknowledged true, and work-notes "Incident <number> raised."
**Step 7: Report**
The new incident number, its priority, the group, and confirmation that the alert now shows the task.
**Rules & Guidelines:**
* Write tools allowed: `servicenow-add-incident` once and `servicenow-update-alert` once, both after the single confirmation. Do not call `servicenow-add-alert`, `servicenow-update-incident` or any other write tool.
* If the add succeeds but the new incident cannot be found in step 6.2, do not create another one. Report the gap and let the user link the alert.
**Validation:**
* If `{alert-number}` is not found, stop.
* If the alert has no configuration-item, create the incident without one and say that the CI is missing.Who it's for
One answer, three teams that use it
Operations engineers
Raise incidents from alerts fast
Turn an alert into a correctly prioritised incident with the right team in one step.
Raise an incident for alert Alert0013544 and assign it to the Database group.
Service desk analysts
Avoid duplicate incidents
Open incidents on the same CI are found first so you link instead of duplicate.
Create an incident for Alert0013602, unless there is already one open for that server.
Monitoring leads
Stop alerts firing again
Every alert is linked and acknowledged once its incident exists.
Turn Alert0013617 into a ServiceNow incident and acknowledge the alert.
How it works
What PopdockAI does when you ask
You give the alert number, and your AI assistant proposes the right incident for it. It is created, linked and the alert acknowledged only after you confirm.
01
Ask
You give the alert number, and optionally a group.
02
Check
It finds the CI's support group and any open incident on the same CI.
03
Confirm
You see the proposed incident with mapped impact and urgency and say yes.
04
Create
It creates the incident, links the alert and acknowledges it.
What PopdockAI provides
The tools behind the answer
Ready made tools for ServiceNow and one secure connection to your AI tool. No code, no exports.
servicenow-get-alert
servicenow-get-configuration-item
servicenow-get-incident
servicenow-add-incident
servicenow-update-alert
This report is built from Get data tools only. PopdockAI applies the row, field and parameter rules your admin sets, and every query is written to the audit trail.
The outcome
Every alert becomes the right incident, once
Related use cases
More ways to put this data to work
One prompt template with infinite possibilities.
Your AI assistant adapts its approach, tone, and capabilities to match each unique context while you maintain a single, manageable workflow.
Join forward-thinking companies that are transforming their operations with intelligent, secure AI automation.

.webp)
