Atlassian Unveils Jira Features for Always-On AI Coding Agents

According to Atlassian, enterprise engineering teams are adopting more agents and are not merely scaling them to do more work. They are running them longer and across more parts of the software development lifecycle. Letting agents operate from start to finish with less tedious supervision carries caveats around trust, grounding, shared context, communication, institutional memory and validation, the company said.

Atlassian is releasing updates that govern agentic activity, set standards for action, review AI agents in motion and validate usage to ensure everything works. The company said agents fail because they lack a good view of project architecture, decisions and standards. Its starting point is Code Context, built with Atlassian's Teamwork Graph, which is designed to give coding agents secure intelligence across complex, multi-repository codebases.

Code Context is combined with Agent Space Settings and Agent Context Controls to govern where agents can operate, what they can see and what they can do. According to Atlassian, this allows teams and management to control agents securely in the same way they control employee access.

Atlassian said the ideal AI agent is not a chatbot. It can run automatically, without iterative prompting, for hours or days on its own, activating when work needs to be done and checking in. To support that model, Atlassian introduced AutoDev, a system that scans backlogs for work and turns it into code merge requests inside Jira. It also introduced a standards system that conforms code and a dedicated agent that reviews merge requests against standards to flag problems. A DevDocs agent automatically generates or updates technical documentation in Confluence directly from code repositories so that documentation does not fall behind.

The company said validation happens alongside transparency. Because there is no perfect playbook or industry standard, and best practices are still being built, Atlassian offers accountability and visibility tools on the back end to track agent activity. Every time agents run, they generate an audit log with diagnostics that allow the system to display and measure AI impact across throughput, quality, adoption and cost.

The tools also include an AI agent usage dashboard that shows who is using the tools, how they are being used and outcomes at a team level. Atlassian said the idea is to help teams move from using only foreground coding tools to working with always-on agents that run in the background and automate engineering tedium. Currently, teams are experimenting with these tools ad hoc but do not have a good library to govern, connect or understand them, according to the company.

Editor's Summary

Atlassian announced upcoming Jira features for always-on AI coding agents, adding context and access controls, automated merge-request and documentation agents, and audit and usage dashboards. The company is positioning the update as a way to help engineering teams move from foreground coding tools to governed, background agents. It also acknowledges that trust, validation and shared context remain central challenges as agent use expands.