Most software teams have automated parts of their release pipeline. The CI/CD runs automatically. Tests fire on merge. But the full pipeline — from the moment a customer reports something to the moment a fix is deployed and everyone who needs to know has been told — still has humans filling the gaps. Someone creates the Jira ticket. Someone updates the status. Someone notifies Slack. These aren't engineering tasks. They're coordination tasks. And they happen every single rele

You've done the hard part, the tests run automatically, the results come in automatically. And then a human has to read them, manually create Jira bugs for every failure, notify Slack, update the release tracker, and write the Confluence summary. The tests are automated. Everything that happens because of the tests is still manual. That's the QA automation gap, and most teams don't notice it until someone asks why QA is still the bottleneck in a team that runs automated tests

MCP, Model Context Protocol, is showing up everywhere in the Atlassian ecosystem right now. But most of the conversation assumes you already know what it is. This post starts from the beginning: what MCP is, why it exists, and why it matters specifically for teams using Jira. The short version: it's the standard that lets AI models talk to your tools, and act on them, without custom integration work for every combination. And it changes a lot about what's possible.

BluBees just launched on the Atlassian Marketplace. Rather than write a post about what we built, we wanted to start with why we built it. Here's something most Jira users recognise but rarely say out loud: Jira doesn't actually run your work. It records it. The real problem is everything that happens in the space between your tools — the manual updates, the missed handoffs, the copy-pasting between Slack and Jira and email. Nobody has solved that from inside Jira. Until now.