AI Briefing
KO

Building a Jira-GitLab Integration with Slack Commands

·2025.11.18 14:23

Key point

A single Slack Command automated Jira and GitLab deployment and merge tasks.

1 / 2

Details

In an environment where tickets are managed in Jira and code is operated in GitLab, every production deployment required humans to repeatedly handle ticket creation, branch/tag management, status changes, and merge tasks, which led to increasing mistakes and omissions. Review requests and deployment notifications were also manual, requiring constant checking to make sure they reached the right people.

To solve this problem, Slack Slash Command was used as the entry point to connect the Jira/GitLab API. An independent Handler was set up for each command, and since most of the work involved I/O operations, asynchronous processing and per-project parallel execution were applied to secure scalability and maintainability.

The implemented features fall into five main categories.

  • Deployment ticket creation: Creates a production deployment ticket and a release branch, and was improved so that if one already exists, it just provides a link instead of creating a duplicate.
  • Pre-deployment validation: Checks via Slack message whether the tag on the Jira ticket matches the latest commit on the release branch, and whether the linked feature branch has been reflected.
  • Merge tasks: Automates feature → develop/release merges and MR creation, reviewer mentions, and the chained release → master → develop merge after production deployment.
  • TEST/STAGE deployment: Bundles ticket lookup, Assignee change, release branch tag creation, and status change into a single command, internally chaining about 7 APIs.
  • Ticket status change: To prevent incorrect changes, built with Slack Interactive Components to go through a preview and confirmation button before performing a batch change.

Through this process, recurring mistakes such as missing tags, missing merges, and incorrect deployment dates were reduced, and pre-deployment validation allowed catching issues before production deployment. As a result, context switching decreased and deployment stability improved, and MR and deployment notifications are now automatically sent as structured messages, improving team-level visibility as well.

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.