One place where humans and agents work under the same rules, off the same context, and in the same flow

Over the last two decades, millions of teams have run their work on Atlassian, writing code, shipping products, and supporting customers across billions of workflows. Our platform is also where they make the hard calls, prioritizing which features to build, deciding which customer problems to chase, and choosing where to invest their resources.

At the center of it all hums the Teamwork Graph, 250B+ connections mapping together the richest context across knowledge, work, people, communication, code, assets, and now, data. Around it sit the apps, services, and agents from Atlassian and beyond that your teams rely on every day. Together, they model your business as it really is, turning all the complexity and nuance of collaboration into a single, governed platform humans and agents can reason across and act on, side by side.

Today, we’re announcing Agentic Multiplayer Protocol (AMP), our foundation for how humans and agents collaborate across the Atlassian platform. It defines how agents take part in multiplayer work with an identity, scoped authority, shared context and tasks, and results you can review.

We’ve always bet on teams. The magic happens when people build on each other’s work, and that doesn’t change when some of your teammates are agents. By grounding agents in enterprise trust, institutional context, and native workflows, AMP unlocks true multiplayer collaboration across your entire System of Work.

Explore new features

Multiplayer only works if you can trust everyone in the loop. AMP gives every agent the context to be useful and the limits to stay safe, with an identity you assign and authority you scope. The Atlassian platform makes it all real. Trusted by 85% of the Fortune 500 and the world’s most heavily regulated industries, it’s designed to protect the full data lifecycle so every new AI capability and agent inherits that same foundation.

  • Atlassian Guard is more than an identity and access management solution. It was built to protect your Atlassian data as ways of working change. Today, that means protecting the data behind AI experiences and agents. Guard Premium goes beyond access to help you discover, classify, and control sensitive information across Atlassian apps. That includes full-site historical scanning to find legacy exposures, real-time scanning across content and attachments, and guardrails across Rovo Chat and connected tools, so the context that makes AI valuable stays secure.
  • Soon, we’re adding controls around what is visible to AI and what it can access with agent accounts and non-human identity inventory. This gives every agent, app, and service account from Atlassian and beyond a full non-human identity so admins can effectively manage autonomous actors from a single place, including revoking access with a simple switch.

With the Atlassian platform, you don’t have to choose between cutting-edge technology and strict compliance. Across regional, industry, and emerging AI frameworks, we continuously strengthen and update our controls to ensure you keep pace of regulatory expectations.

  • We’re introducing local EU AI inference, which restricts LLM processing exclusively to models hosted within the EU, so you can run your most sensitive workloads in a region you trust.
  • We meet you at your regulatory bar with proactive compliance for regional and AI-specific standards like the EU AI Act and ISO 42001 for Rovo, all reinforced by audit logging and observability.

Compounding context the whole team can count on

Models are intelligence you can rent, and providers make them better by the week. But what they don’t have is the wisdom of how your business actually works. That’s what the Teamwork Graph captures. It lives where the work happens, an institutional memory fed by every closed ticket and documented decision. Under AMP, every agent draws on it, whether it’s working inside or outside Atlassian. And each time your multiplayer teams collaborate on a page, update a work item, or resolve an incident, the context compounds and the next interaction gets even better.

“Atlassian provides the context we need to make AI work for us. We’ve proven that a good model powered by Atlassian’s organizational context will always outperform the best model in the world working in a vacuum.”

Jose Ramirez
Executive Director, Enterprise Agility & Delivery, Kaplan north america

The Teamwork Graph is already your edge. Today, we’re making it even sharper:

  • Rovo Code Search connects your codebase to your business context. The Teamwork Graph now reads your source code itself, down to functions, symbols, and classes. The new Rovo Code Search app gives engineers one place to search and navigate all of it across Bitbucket and GitHub, in plain language, without cloning a repo. Their coding agents draw on that same context, so they work with a real understanding of the codebase and burn fewer tokens getting there.
  • Data context puts your data lakes and warehouses to work, and Atlassian Insights is where you use it. Structured data is now a Teamwork Graph context type, so you can query your lakes and warehouses in plain language wherever you work. Atlassian Insights is the new dedicated home for that data context, curating 20+ structured sources with catalog, glossary, and lineage, so anyone can pull a quick answer or create lasting insights for their team.
  • Connectors bring the rest of your world in. We’ve expanded our library of 80+ Connectors with additions like Zoom, Gong, Microsoft Entra ID, and Google Identity, so you can pull in context from the tools you use every day for your agents to act on. Plus, we’ve shipped 65+ improvements to make them even more secure and capable.
  • Agent sessions and agentic updates in Focus make agent work visible. With agent sessions in Jira, local and cloud agent runs automatically link back to work items, making autonomous activity visible, traceable, and reusable across the team. At the portfolio level, agentic updates in Focus synthesize cross-project progress and goal signals into real-time summaries, helping leaders spot risks and blockers without manual reporting.

Humans and agents, one loop

On the Atlassian platform, work moves through a loop your teams already know: something gets created, it gets reviewed and commented on, and that triggers the next action. What’s new is who’s in the loop. Today, humans and agents already collaborate over ten million times a month across the Atlassian platform. With AMP, agents work alongside teammates just like… teammates. They join the conversation, act on shared context, and hand back results you can review. Here are just some of the new products and experiences you get with AMP across the platform:

  • Rovo Work takes on the complex stuff. Work is a new mode in Rovo Chat that handles complex, multi-step tasks with human oversight to review and approve. You start with a goal, it proposes a plan, you shape it, and it executes across Jira, Confluence, and your connected tools. A task can run for hours, looping until it finds the right tools, teaches itself new ones, runs scripts, and completes the ask, all in a secure sandbox your admins govern.
  • Artifacts app gives any agent’s work a durable home. AI-generated plans, interactive models, and HTML views no longer get trapped in chat windows or lost in downloads. The new Artifacts app gives AI creations from any AI tool – on or off Atlassian – permanent URLs that embed cleanly into Confluence pages, Jira work items, or Slack channels, all governed by enterprise permissions and immediately indexed back into the graph.
  • The loop doesn’t stop at Atlassian’s edge. Atlassian MCP is one of the most used in the world, with millions of tool calls every day. It means the create → review → act loop runs wherever your people and agents work, all drawing on the same context and writing every step back to the Teamwork Graph. We rebuilt it from the ground up so any agent can join the loop, going from dozens of tools to 200+ across Atlassian and beyond.

“Atlassian holds our company’s institutional memory. By using the MCP with Rovo, we can securely plug that entire context layer into our everyday AI workflows. Questions that used to take days of alignment across teams are now answered in seconds, allowing us to unlock organizational knowledge wherever our teams are working.”

Tanner Randolph
EVP, Global CIO & CISO, Applied Systems

Unleash the potential of your multiplayer teams

Atlassian was built on the conviction that teamwork is how great things get done.

Over the years, we’ve broken down silos across tools and teams with open APIs, industry-leading MCPs, and a Teamwork Graph that meets people where they work. Because we know solo features, protocols, and individuals can only get so far, and the real leap happens when it all connects.

That’s what AMP delivers. A collection of experiences, technologies, and patterns focused on helping people and agents have a shared operating model. Everything your team builds adds up instead of fragmenting.

It’s live across the Atlassian platform today. Your whole team is in the loop now, humans and agents both.

Explore more at Team ’26 Europe

We announced these updates at Team ’26 Europe, our annual user conference, along with the rest of this year’s news. The sessions are available live and on demand if you want to go deeper.