AI AGENTS

Post-Incident Postmortem Draft on PagerDuty Resolve

When a PagerDuty incident resolves, an agent assembles the timeline from PagerDuty and Slack, drafts a blameless postmortem.

CategoryAI Agents
Enginepaperclip
Difficultyintermediate
Triggerevent
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerPagerDuty incident resolvedPagerDutyPagerDuty
  • LogicFilter to postmortem-worthy incidents
  • ActionCollect timeline + Slack threadSlack
  • ActionAgent drafts blameless postmortem
  • ActionFile draft Confluence pageConfluenceConfluence
  • OutputLink draft to commander in SlackSlack

What it does

Automates the tedious first draft of a postmortem. As soon as an incident is resolved, the agent pulls the incident timeline, the responder Slack thread, and the metric context, then writes a blameless postmortem covering impact, timeline, root cause, and follow-up actions. It files the draft for human editing rather than publishing.

When to use it

Use it after any significant incident so the writeup starts at 80% instead of a blank page. The incident commander refines and approves; the agent handles the grind of stitching the narrative together.

How it works

  1. 1A PagerDuty incident resolution triggers the flow.
  2. 2A logic step filters to incidents above a severity or duration threshold worth a postmortem.
  3. 3The agent collects the incident timeline and responder messages from the Slack channel.
  4. 4It drafts a structured blameless postmortem with timeline, root cause, and action items.
  5. 5The draft is filed as an unpublished Confluence page.
  6. 6A Slack message links the draft and asks the incident commander to review and publish.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect PagerDutyIncidents, on-call, escalations.
  2. 2
    Connect SlackChannels, DMs, threads, mentions.
  3. 3
    Connect ConfluenceSpaces, pages, blueprints.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.