Atlassian evolves Jira into an orchestration hub for developers and AI agents

Atlassian Corp. today announced it is expanding Jira with updates that will help developers prepare, distribute and track work performed by artificial intelligence agents.

The company’s new Jira Planner helps turn incomplete project ideas into technical specifications, while its Jira Coding Agent and integrations with third-party agents transform work items into requests. With automation rules and an agentic engineering template, teams can design workflows where agents are assigned work as projects move through the platform.

The idea behind the updates is to position Jira as the place where developers go to organize and prepare, dovetailing with the agentic AI coding boom. Coding-agent adoption is already widespread, but Atlassian sees the resulting improvements beginning to plateau as planning and coordination become a bottleneck.

“We need a solution rather than a tool,” said Head of Engineering for DevAI Ming Wu. “We need a solution across the entire software development lifecycle journey to actually address different pain points.”

This is where Jira’s history becomes the connective tissue of software team progress and processes become strategically important. The company wants to become the control plane for a mixed workforce of developers and agents, whether those agents run locally, in the cloud or from third parties.

“The more we talk to our customers, based on our DevEx report and also DX survey, the more we look at data, the more we believe the customers need some holistic solution to pull those AI tools together and realize the gain — the return on investment — from their AI tool usage,” Wu said.

Even with the broad adoption of coding tools, the market is in a transition. Even as developers become more comfortable using an expanding range of AI tools, companies are struggling to find reliable, company-wide value.

The answer from Atlassian has been to broaden its audience and expand its tool capabilities. Instead of measuring success by how rapidly an agent produces code, Jira can help address delays caused by unclear requirements, missing project context, handoffs, environment setup, assignment, documentation, review and governance. Essentially, this is the “work that surrounds work.”

Industry research has shown that even as frontier models get smarter and faster, submissions from AI agents are accepted less frequently than human-authored ones and tend to be structurally simpler. According to a study by researchers at Queen’s University Kingston in Canada, a review of 61,000 repositories and 47,000 developers found that AI agents aren’t even second-class citizens; they’re closer to a carefully tended third-class. AI agents may produce a great deal of code quickly, but the quality is considered extremely low.

Getting agents to do good work requires giving them a way to coordinate, collaborate and augment human developers, who do the higher-quality work. However, if humans spend all their time orchestrating, managing and fine-tuning agents for small, laborious and boring tasks, they’re not doing their own high-quality work.

To help with that, Jira is adding a way to easily launch a ticketing board system that allows agents to enter into the collaborative flow called the Agentic Engineering Template. It’s for more advanced adopters who want Jira workflows to work alongside agents.

“It helps you to set up the Jira board where columns have a state,” Wu said. “When you move the issue from one column to the other, agents are automatically assigned, so there’s a certain level of more detailed, nuanced automation in that feature.”

Jira Coding Agent handles most bounded work in the cloud, meaning developers don’t need to use a coding editor to trigger it. It can act directly from Jira, but it’s not a complete or direct replacement for hands-on development. It exists so developers can delegate tasks they’d rather get out of the way without interrupting their current work.

The company also announced that work can be assigned to any coding agent. This means developers can still use their favorite agentic environment for issuing requests, including Claude, Codex, Cursor, or GitHub Copilot, directly from Jira. The work stays grounded in Jira’s knowledge and history and the project’s path, and it feeds critical information to the agent so it can get the job done.

“You want to bring the true value rather than boosting usage itself,” Wu said. “[Boosting] usage itself is superficial, really. You want the customer getting good value for their money spent.”

Image: SiliconANGLE/Microsoft Designer

A message from John Furrier, co-founder of SiliconANGLE:

Support our mission to keep content open and free by engaging with theCUBE community. Join theCUBE’s Alumni Trust Network, where technology leaders connect, share intelligence and create opportunities.

15M+ viewers of theCUBE videos, powering conversations across AI, cloud, cybersecurity and more

11.4k+ theCUBE alumni — Connect with more than 11,400 tech and business leaders shaping the future through a unique trusted-based network.

About SiliconANGLE Media