/
/
Raise one ServiceNow problem for incidents with a shared cause
For problem managers who want related ServiceNow incidents covered by one problem and an analysis task, by asking their AI assistant.
On this page
Connector
Connect ServiceNow, pick your AI tool and paste the prompt.
The problem
Shared causes with no problem record, or two
When several incidents share a cause, someone needs to raise one problem and start a root cause investigation. By hand, the link is not checked, an existing problem is missed and duplicated, or the analysis task never gets assigned.
Without PopdockAI
- Read each incident to see if they are related
- Miss an existing problem that covers them
- Create the problem from a blank form
- Forget the root cause analysis task
With PopdockAI
- Each incident is read and a real shared cause is checked
- Existing open problems that cover them stop the job
- Symptom, impact, urgency and group are proposed for you
- The problem and analysis task are created after you confirm
The prompt
Use this prompt
Copy it into any AI tool connected to PopdockAI. Swap the values in braces for your own.
**Objective:** Raise one problem covering incidents `{incident-numbers}`, owned by `{assignment-group}` (default: the group most of the incidents belong to), with a root cause analysis task.
**Step 1: Read the incidents**
Call `servicenow-get-incident` with number set to each incident in `{incident-numbers}`. Collect priority, configuration-item, service, category, assignment-group and short-description.
**Step 2: Check they belong together**
If the incidents share no configuration-item, service or clear symptom, say so and ask the user to confirm they are related before going on.
**Step 3: Check for an existing problem**
Call `servicenow-get-problem` with short-description.contains set to the shared CI name or key words. If an open problem already covers them, report its number and stop.
**Step 4: Draft and confirm**
Propose:
* short-description: the shared symptom and CI in one line
* description: the list of incident numbers with their opened dates and short-descriptions
* configuration-item and business-service from the incidents when they agree
* impact and urgency: the highest seen across the incidents
* assignment-group: `{assignment-group}` or the most common group
Show this and ask for an explicit yes.
**Step 5: Write, in this order**
1. Call `servicenow-add-problem` once with the confirmed values.
2. Call `servicenow-get-problem` with short-description.contains set to the new short-description and opened.from today to get its number and sys-id.
3. Call `servicenow-add-problem-task` once with problem set to that sys-id, type Root cause analysis if listed, the same assignment-group, and short-description "Root cause analysis for <problem number>".
**Step 6: Report**
The problem number, the RCA task number (read it back with `servicenow-get-problem-task` using problem set to the problem number), and the incidents it covers.
**Rules & Guidelines:**
* Write tools allowed: `servicenow-add-problem` once and `servicenow-add-problem-task` once, after one confirmation. Do not call `servicenow-update-incident` or `servicenow-update-problem`; this tool set cannot link incidents to a problem, so list the incident numbers in the description instead and say so in the report.
* If step 5.2 cannot find the new problem, stop and report; never create a second one.
**Validation:**
* If any incident number is not found, list it and ask whether to continue without it.Who it's for
One answer, three teams that use it
Problem managers
Open problems from incident patterns
Raise one problem with a root cause task, with duplicates ruled out first.
Raise a problem for INC0020871, INC0020902 and INC0020944, owned by the Database group.
Service desk leads
Escalate repeat issues properly
Turn a set of similar incidents into a problem without checking for duplicates by hand.
These incidents look related: INC0021015, INC0021031. Raise a ServiceNow problem if none exists.
Infrastructure engineers
Start the investigation formally
Get a problem and analysis task in place so the root cause work is tracked.
Open a problem for the three VPN incidents this week, INC0021100, INC0021112 and INC0021130.
How it works
What PopdockAI does when you ask
You give the incident numbers, and your AI assistant checks they belong together and are not already covered. The problem and its analysis task are created only after you confirm.
01
Ask
You give the incident numbers, and optionally an owning group.
02
Check
It confirms the incidents share a cause and no open problem covers them.
03
Confirm
You review the proposed problem with symptom, impact and group, and say yes.
04
Create
It creates the problem and a root cause analysis task and reads it back.
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-incident
servicenow-get-problem
servicenow-add-problem
servicenow-add-problem-task
servicenow-get-problem-task
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
One problem with an investigation already assigned
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)
