Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for it support specialists

Support Incident Correlation Agent

Related tickets are grouped into the correct incident, quickly and with evidence.

AnalysisAutomationCustomer supportOperationsIT and development
See how it worksGet access
Support Incident Correlation Agent: what goes in, what the agent does and what you get

What it does

When an outage starts, many customers report it in different words, and support agents investigate the same problem separately. Customers get mixed answers and engineering gets duplicate pings. This agent clusters new tickets by symptom and affected service, queries telemetry for those services, and checks whether the evidence supports one shared cause. When telemetry shows two different failing services, it splits the cluster so each problem reaches the right team. It keeps one internal incident candidate per cause, with linked tickets, affected customer counts and the latest telemetry. Only the incident coordinator declares an incident and approves customer replies. Edge case: tickets that match the symptom but come from a region with healthy telemetry stay outside the cluster and are reviewed separately.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedNo 1 STARTS WHEN Ticket spike or outage-like ticket 2 USES A TOOL Cluster tickets by symptom and service 3 USES A TOOL Query telemetry for the affected services 4 CHECKS THE RESULT Does telemetry support one shared cause? If not: split the cluster by service. Back to step 2. 5 DOES Create or update the internal incident candidatewith links 6 YOU APPROVE Incident coordinator declares the incident andapproves replies 7 RESULT Incident candidate maintained until resolved
Read the steps as a list
  1. Ticket spike or outage-like ticket
  2. Cluster tickets by symptom and service
  3. Query telemetry for the affected services
  4. Does telemetry support one shared cause?If not: split the cluster by service. Back to step 2.
  5. Create or update the internal incident candidate with links
  6. Incident coordinator declares the incident and approves repliesThe agent waits here for your OK.
  7. Incident candidate maintained until resolved

How it decides

For each new ticket it decides 'join an existing candidate' or 'start a new one' by symptom similarity and affected service, then confirms with telemetry; contradictory telemetry splits a group.

  • Join or new: by similarity of symptom and affected service.
  • Split: when telemetry shows different failing services.
  • Escalate to declared incident: only a coordinator decides.

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Ticket spike that starts clustering (default 10 similar tickets in 15 minutes)
  • Similarity threshold for joining a cluster (default 80%)
  • Telemetry sources queried (default service health, error rates, identity provider status)
  • Who declares incidents and approves replies (default the on-call incident coordinator)
  • Update interval for incident candidates (default every 5 minutes)

What keeps you in control

It always asks you first

  • Declaring a public incident
  • Status page updates
  • Mass replies to customers

Hard limits

  • Internal board only; no public statements.

It stops when

  • Done: incident resolved and all tickets linked.
  • Nothing to do: no cluster above threshold.

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensBetween 9:05 and 9:20 on May 14, twenty tickets say 'can't log in'. Telemetry shows the login service healthy, so the one-cause check fails. The agent splits the cluster: 12 tickets trace to single sign-on failing for one identity provider, 8 to delayed verification emails. It creates two incident candidates. Coordinator Victor Hale declares the sign-on incident and approves a status reply.

More agents for it support specialists

Quality control · Automation

Support Knowledge Repair Agent

Help articles answer real customer questions correctly, verified against resolved tickets.

8 steps1 check4 jobs
Customer service · Customer support

Support Device Compatibility Agent

Troubleshooting that fits the customer's actual hardware combination.

8 steps1 check5 jobs
Customer service · Customer support

Support Incident Follow-up Agent

Every affected customer is verified fixed before their ticket is closed.

8 steps1 check7 jobs
Communication · Customer support

Outage Surge Communication Agent

Give agents and customers one accurate message during an outage and keep it up to date.

11 steps2 checks9 jobs