A scheduled automation that turns a week of noisy support signals from Slack, Asana, and WhatsApp into a structured QA digest, posted to the team every Monday.

Support signals arrive across multiple channels: Slack threads, Asana form submissions, and WhatsApp group chats. Without aggregation, patterns get lost and the QA team spends time manually triaging volume rather than acting on it.

This tool runs every Monday at 04:00 UTC via GitHub Actions, pulling the past seven days of data, deduplicating issues across sources, and using GPT-4o to synthesise a structured summary. The report covers issue volume, software vs hardware breakdowns, service requests, and recommended QA actions for the week. Severity-1 issues trigger an immediate alert to the team channel.

A 12-week rolling history window lets the model flag recurrence patterns, comparing the current week against past data to surface issues that keep coming back.

Aggregates support data from Slack (#product-support), Asana (Product Support form), and WhatsApp exports

Deduplicates issues across sources to avoid double-counting the same report

Uses GPT-4o to synthesise a structured summary with volume stats, issue categorisation, and actionable QA recommendations

Posts the weekly digest to #team-qa automatically every Monday at 04:00 UTC

Triggers severity-1 alerts when critical issues are detected in the aggregated data

Tracks recurrence over a 12-week rolling window and flags recurring issues explicitly

Fully automated via GitHub Actions with zero manual intervention required

Python 3.11

OpenAI GPT-4o

Slack API

Asana API

GitHub Actions