Post-incident Review Agent · Use case
How can a telecom operator use an AI agent to prepare a post-incident review on its own infrastructure?
A post-incident review agent sets out what happened in an incident from the trouble tickets, the alarms and the notes taken on the bridge. It lists the decisions that were made and the gaps between them, and drafts the review in the operator's own format. The incident manager checks every line and owns the findings and the actions. The agent names no cause the records do not show.
The outage lasted an hour.
The review takes a fortnight.
- ✗ 01The timeline is spread across tickets, alarm logs and a chat, and one engineer rebuilds it by hand after the event.
- ✗ 02By the time it is written, the people on the bridge have moved to the next incident.
Where it stops for a person
Always. The incident manager owns the findings.Always. The incident manager decides the cause and assigns the actions. Where the records are silent the agent says so.
It builds the timeline.
The incident manager owns the findings.
Step · 01
The incident manager hands the agent the incident's reference and the bridge notes.
Step · 02
It reads the tickets and the alarms for the period and puts every event in order.
Step · 03
It marks each decision, who made it and what was known at the time.
Step · 04
It lists the gaps: stretches with no recorded action, and events the records disagree about.
Step · 05
It drafts the review in your format, each line with its source. The incident manager corrects it and writes the findings.
Reads from · connectors
Your own APIs
File Upload
Tickets and alarms, through your systems' own APIs; where they follow TM Forum Open APIs, the descriptions are already public. Bridge notes are uploaded files. Your review format is a knowledge source.
Acts through · MCP servers
Nothing. It changes no ticket and sends nothing.
What you set up
- Connect the tickets and the alarms with read-only access.
- Index your review format.
- Test against incidents whose reviews you have already written.
Where it runs
The agent, its knowledge store and the model run on infrastructure you own or control. You choose the model and serve it yourself, so questions, documents and answers are not sent to a public AI service. Incident records and network data stay inside the operator.
How you get it today
This page is labelled Use case: the agent is built in Katonic Studio. You describe the job, connect what is named above, add test questions and release the agent once it passes them. When the Catalog holds a tested template for this job, this page will carry the Template label and show how to take it.
Does the agent say what caused the incident?
No. It sets out what the records show and where they are silent. The cause is the incident manager's finding.
Can it prepare the report our regulator requires?
It prepares the timeline that report needs. Your named officer decides whether to report, and submits it.
What about what was said on the bridge and never written down?
It is not in the review unless someone adds it. The draft lists the gaps.