Atlassian 升级 AI 编码代理，面向“always-on”软件开发

UPDATED 12:00 EDT / SEPTEMBER 10 2026 AI Atlassian upgrades AI coding agents for always-on software development by Kyt Dotson Identifying the next phase of software development as always-on agentic artificial intelligence, Atlassian plc today announced a new set of upcoming Jira features to help engineering teams run AI agents at large scale while governing actions over long periods. As enterprise engineering teams begin to adopt more agents, they’re not just scaling agents to do more work – they’re working them longer and across more parts of the software development lifecycle. According to Atlassian, letting agents work with less tedious supervision from start to finish comes with caveats because of trust, grounding, shared context, communication, institutional memory and validation. To address this, the company is releasing updates that govern agentic activity, set standards for action, review AI agents in motion, and validate usage to ensure everything works. According to Atlassian, agents fail because they don’t have a good view of the architecture of projects, decisions and standards. So, the starting line grounds them with better context, and the company calls this Code Context. It’s built with the company’s Teamwork Graph, which gives coding agents secure intelligence across complex, multi-repository codebases. This 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. Allowing a team and management to securely control them the same way they control employee access. The ideal of an AI agent is that it’s not a chatbot. It can run automatically, no more prompting it iteratively. This means that it can run for hours or days on its own, activating when work needs to be done and checking in. Atlassian has introduced AutoDev, a system that scans backlogs for work and turns it into code merge requests inside Jira, a standards system that conforms code and a dedicated agent that reviews merge requests against standards to flag problems. Finally, when everything is rolling, a DevDocs agent automatically generates or updates technical documentation in Confluence directly from code repositories so that documentation doesn’t fall behind. Automatic but not invisible All of this doesn’t happen in the dark; validation happens along with transparency. Because there isn’t a perfect playbook or industry standard, but best practices are still being built, Atlassian offers accountability and visibility tools on the back end to track it. 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. It lets teams understand spend across software as it ships. It also includes an AI agent usage dashboard that shows who is using the tools, how they’re 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 pick up the slack by running in the background to automate engineering tedium. Currently, teams are experimenting with these tools ad hoc, but don’t have a good library to govern, connect or understand them. 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 Are you an AWS customer?  Support SiliconANGLE financially by buying your AWS services from our Marketplace portal page and links: https://siliconangle.com/aws-marketplace/ About SiliconANGLE Media SiliconANGLE Media is a recognized leader in digital media innovation, uniting breakthrough technology, strategic insights and real-time audience engagement. As the parent company of SiliconANGLE, theCUBE Network, theCUBE Research, CUBE365, theCUBE AI and theCUBE SuperStudios — with flagship locations in Silicon Valley and the New York Stock Exchange — SiliconANGLE Media operates at the intersection of media, technology and AI. Founded by tech visionaries John Furrier and Dave Vellante, SiliconANGLE Media has built a dynamic ecosystem of industry-leading digital media brands that reach 15+ million elite tech professionals. Our new proprietary theCUBE AI Video Cloud is breaking ground in audience interaction, leveraging theCUBEai.com neural network to help technology companies make data-driven decisions and stay at the forefront of industry conversations. LATEST STORIES Nasdaq invests $100M in Kraken parent company Payward to advance tokenized equities Typewise orchestrates customer-service AI agents Gravwell adds five AI agents that gather their own investigation context Exclusive: Cfo.ai launches an agentic CFO for business founders Pigment launches AI-generated interfaces to broaden business planning CoreWeave launches new engineering service to help enterprises implement physical AI LATEST STORIES Nasdaq invests $100M in Kraken parent company Payward to advance tokenized equities BLOCKCHAIN - BY KYT DOTSON . 30 MINUTES AGO Typewise orchestrates customer-service AI agents AI - BY PAUL GILLIN . 44 MINUTES AGO Gravwell adds five AI agents that gather their own investigation context SECURITY - BY DUNCAN RILEY . 3 HOURS AGO Exclusive: Cfo.ai launches an agentic CFO for business founders AI - BY MIKE WHEATLEY . 3 HOURS AGO Pigment launches AI-generated interfaces to broaden business planning APPS - BY PAUL GILLIN . 3 HOURS AGO CoreWeave launches new engineering service to help enterprises implement physical AI AI - BY MIKE WHEATLEY . 3 HOURS AGO