PROJECT MANAGEMENT

Retro Blocker Triage to GitHub Issues and PagerDuty

Parses a Zoom retro transcript, separates engineering blockers from general action items, opens GitHub issues for the blockers.

CategoryProject Management
Enginesim
Difficultyadvanced
Triggerevent
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerZoom retro recording completedZoomZoom
  • ActionFetch recording transcriptZoomZoom
  • ActionClassify blockers vs action items with severityOpenAI
  • LogicBranch on blocker severity
  • ActionOpen labeled GitHub issues for blockersGitHubGitHub
  • OutputPage on-call for critical blockersPagerDutyPagerDuty

What it does

This workflow reads a completed Zoom retro, classifies each surfaced item as either a routine action item or an engineering blocker, and routes them differently. Blockers become GitHub issues in the relevant repo with severity labels, and any blocker the team called out as actively breaking work triggers a PagerDuty alert so it does not wait for the next standup.

When to use it

Use it when retros regularly surface real engineering blockers that need tracking in GitHub and, occasionally, immediate escalation. It keeps low-severity follow-ups as ordinary issues while ensuring a critical blocker gets a human paged the same day.

How it works

  1. 1A Zoom recording-completed event fires for the retro.
  2. 2The transcript is fetched from Zoom.
  3. 3OpenAI classifies items into blockers versus action items and tags each blocker with a severity.
  4. 4A logic step branches on severity.
  5. 5Every blocker is opened as a labeled GitHub issue.
  6. 6Critical blockers additionally trigger a PagerDuty incident routed to the on-call engineer.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect ZoomMeetings, recordings, transcripts.
  2. 2
    Connect OpenAIModels, embeddings, files.
  3. 3
    Connect GitHubRepos, issues, pull requests, actions.
  4. 4
    Connect PagerDutyIncidents, on-call, escalations.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    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.