PagerDuty, answered where the incident is discussed.
Ask what is open, who is on call and what happened on the timeline, and Cedric proposes the acknowledge, the note or the override before anything pages a person.
100K credits ($50) on sign up. No card needed.
- During an incident the first questions are what is breaking, who has it, and what has been tried, and PagerDuty holds all three.
- Ask in plain words: open incidents by service, team, urgency, status or time range, with each one's timeline, notes and related alerts.
- Cedric can acknowledge, resolve, escalate, reassign, or change an incident's priority and urgency, and add the note or status update responders need.
- He looks up who is on call now, reads schedules and escalation policies, and creates a schedule override when someone needs cover.
- Triggering, resolving or reassigning an incident pages real people, so each one waits for your yes.
What can a PagerDuty AI agent do?
The useful part of PagerDuty during an incident is the record it keeps, and the useful part after is the same record. A PagerDuty AI agent reads it from the conversation. Cedric lists and filters incidents by service, team, urgency, status or time range to show what is breaking right now, and reads a single incident's full details: the timeline of events, the notes responders left, and the alerts attached to it. He also browses services, teams and users, including each person's contact methods and notification rules, so "who gets paged for the payments service" has an answer.
Acting on an incident is proposed first. Cedric can acknowledge, resolve, escalate or reassign an incident, or change its priority and urgency, and he can add a note or a status update so responders and the people waiting on them see the same thing. He can also trigger a new incident on a chosen service to page the on-call responder when something is found outside the alerting. Each of those writes reaches a real person's phone, which is why they wait for your yes. That is the shape of an operations assistant for the team that carries the pager.
The rest is coverage. Cedric looks up who is currently on call, reads schedules and escalation policies, and creates a schedule override to cover a shift when someone is out. He creates or lists maintenance windows so planned work does not page anyone. PagerDuty enforces request rate limits per key, so a very wide listing is paged rather than pulled in one go.
Mid-incident
Who has it, what was tried, and one status update.
Marcus Whitfield9:14 AM
@Cedric what's open on the checkout service right now, who's on call, and post a status update saying we've rolled back the 14:10 deploy and are watching error rates.
CedricAPP9:15 AM
One open incident on Checkout: #2291 "5xx rate above 5%", high urgency, triggered 14:22, acknowledged by Ines Laurent at 14:24. Timeline so far: two related alerts, one note from Ines ("rolling back 14:10 deploy"). On call for Checkout is Ines until 18:00, then Tom Bakker. Proposed status update on #2291: "14:10 deploy rolled back at 14:31, watching error rates before resolving." Nothing is posted yet.
Looked up
Open incidents by service, team, urgency, status or time range, with each one's timeline, notes and related alerts.
Handled
Acknowledged, resolved, escalated, reassigned, or given a new priority, urgency, note or status update, proposed first.
Covered
Who is on call now, the schedule and escalation policy behind it, and an override or maintenance window when one is needed.
Connect PagerDuty with your own key
No setup project, no workflow builder, no field mapping. The flow is the same for every tool Cedric connects to.
- 01
Install Cedric where your team talks
One click. He joins your Slack like any other member, and everyone shares the same Cedric.
- 02
Connect PagerDuty from the dashboard
Open the Integrations page, pick PagerDuty and paste an API key from PagerDuty's integration settings. You can revoke the connection in one click at any time.
Pick Read-only when you connect if you only want him reading. You can switch it later.
- 03
Ask in plain words
Back in the channel, try it: "what's open right now and who is on call?" You describe the outcome; Cedric picks the PagerDuty actions the ask needs. There is nothing to configure.
Connect for the whole workspace so the team shares it, or keep the connection personal. Two accounts are fine too; each gets a short name.
- 04
Let it become routine
Schedule the recurring version of the ask, and let Cedric fold PagerDuty into bigger jobs that cross your other tools. Anything that would leave your workspace still comes back as a preview to approve first.
Better in company
An incident usually starts in the product and ends in code. Cedric can read the PagerDuty timeline, file the follow-up in Linear or open the issue in GitHub with the incident linked, and note on the incident that the fix is tracked, each step on your yes.
The postmortem lives in Notion and the retro call happens in Google Meet. Ask Cedric for last week's incidents by service and he can draft the timeline section from the notes and alerts PagerDuty already holds.
Worth knowing
PagerDuty comes from Cedric's catalog of 3,200+ apps. He works from what the PagerDuty connection can read and do, the same list its card shows in your dashboard.
Writes page real people. Triggering a new incident, resolving one, or reassigning it reaches someone's phone right away, so Cedric proposes the action and asks every time.
PagerDuty's per-key rate limits apply. A wide listing across many services or a long time range is paged rather than pulled in one go.
The PagerDuty connection sets the scope. Cedric sees what it grants. Access is logged, and you can revoke it from the dashboard at any time.
Nothing sends itself. An email, a post, a change to a record: anything with consequences comes back as a preview for a person to approve.
FAQ
Yes. PagerDuty is one of the 3,200+ apps in Cedric's catalog. Connect it once from the dashboard, with an API key from PagerDuty's integration settings, and ask in plain words. Cedric picks the actions an ask needs from what the connection can read and do, with no fields to map.
Yes, by triggering a new incident on a chosen service, which pages that service's on-call responder through the escalation policy. Because it reaches a real person, Cedric shows the service and the incident title and waits for your yes before triggering.
He can create a schedule override to cover a shift, which is how a swap or a sick day is handled in PagerDuty. He reads schedules and escalation policies to show who is on now and next, and proposes the override with the exact window before saving it.
Connect it for the whole workspace and everyone shares it, with one person authorising once. A connection you would rather not share, a personal inbox for instance, can stay personal to you.
Yes, and it starts that way: a key connection is read-only until you turn writes on. Even with writes on, anything that changes or sends waits for an approval.
Every workspace starts with 100,000 free credits (worth $50) on sign-up, no card. Plans start at $19 a month for the whole team, no seats. Each task uses credits in proportion to the work. See pricing.
Works well with PagerDuty
AWSServers, storage, logs and the bill read from the thread; restarts and resizes on your yes.
CloudflareDNS, cache and firewall read and adjusted from the thread, every live change on your yes.
GitLabIssues, merge requests and pipelines read and moved from the thread, every merge on your yes.
SentryErrors found, stack traces read, issues triaged and assigned from the thread, on your yes.
DatadogMetrics, logs and monitors read in the thread; alerts muted and monitors changed on your yes.
AsanaAssignees asked, tasks updated from their answers, and the project current before every meeting.
Connect PagerDuty in a minute.
Ask what is open and who is on call, then see the status update he drafts.
100K credits ($50) on sign up. No card needed.