An agent is only as good as the context you give it. Feed it the right pages, recordings, relationships, decisions, and chat history, keep that context accurate and current, and it ships real work. Cut it off or feed it stale data, and it hands you confident nonsense.
The Teamwork Collection, Jira, Confluence, Loom, and Rovo, is built on a shared context layer called the Teamwork Graph. It gives every agent and teammate direct access to the same rich, always-current context, right inside the tools your teams already use. If you’re new to the Collection, that graph is the foundation you start building the moment you work. If you’re already in it, it’s the advantage you keep compounding.
At Team ’26 Europe, we’re building on that foundation across three areas: Context, Collaboration, and Confidence. Give your AI richer context to draw on, let your people and agents act on it together, and stay confident about what’s happening as work speeds up. Agents do the legwork, you make the calls. Here’s what’s shipping.
Context: a richer, connected foundation
Your AI is only as sharp as the context behind it. These updates pull more of your work into the Teamwork Graph and keep it current, so every agent and teammate draws on the same complete picture.
Confluence + Loom MCP tools and skills
You can now create and edit Confluence pages and pull context from Loom videos using MCP tools and skills, extending what the Teamwork Graph already knows. This means agents can run on their own and can hand work to each other. Tell one to “use this Loom to draft release notes in Confluence,” and it does the whole thing without you having to step in at each stage. You set the brief and sign off on the result; the agent handles the steps in between. It all runs on the new and updated Atlassian MCP, our next-generation protocol.
Jira app for Slack and Microsoft Teams
The Jira app for Slack and Microsoft Teams turns a conversation into tracked work. Route a context-rich thread straight to the right agent or teammate in Jira, so decisions made in chat don’t evaporate when the channel moves on.
Third-Party Folders in Confluence
New folders in Confluence sync content from other SaaS apps. Point one at your Google Drive sales decks, and they show up in Confluence and stay current. No copy-paste, no reconciling which version is the latest.
Great work happens between people and agents, not alone. These updates help your team and your agents create, hand off, and act on work together.
Collaboration: people and agents, together
Great work happens between people and agents, not alone. These updates help your team and your agents create, hand off, and act on work together.
Record for Agent
Record for Agent, now in open beta, turns a quick Loom video into something an agent can act on. Talk through what you want and show it on the screen; Loom captures your thoughts and feedback as an audio and visual brief, and the agent picks up the rich details and nuance and takes action from there, like creating Jira work items, a prototype, or even slides in Confluence.
Third-Party Agent Triggers and Automations
Bring your own agents into Jira and Confluence. Run third-party agents alongside Atlassian’s agents, add them to your workflows, and keep a full history of what each did and when. A status change in Salesforce can now start an agent workflow on its own. The follow-up doc or task gets created when the trigger fires, not when someone finally remembers to do it. You define the trigger and where the workflow needs a human check before anything goes out; the agent just doesn’t forget the step.
Tell a richer story with AI overlays in Loom
Now in early access, Loom automatically adds visual graphics such as animated titles, key-point highlights, and intro or outro cards to your recording, giving you a richer, more polished video. These recordings tell a more nuanced story in a format that is easy to share with your team.
Artifacts app and smart links in Confluence
Everything your AI creates now has one home. The Artifacts app is where you find any AI-created doc, draft, or asset, and smart links surface them across Confluence. Instead of dying in a one-off agent session, that output becomes shared, reusable context that lives in Confluence and feeds back into the Teamwork Graph, so your team and your next agent can build on it.
Confidence: see through the noise
As agents multiply output, speed can turn into noise. These updates surface the signal from across your work, so managers and teams can act before small problems become big ones.
Individual Capacity Planning
New capacity views in Jira show how work is distributed across a team. Managers and team leads can see who’s overloaded and who has room, and rebalance before burnout or slipped deadlines set in.
Project Budgeting
Jira now calculates project signals for you, budget variance, risk scores, days until a deadline, so risks surface early and teams can course-correct while it still matters.
Get started
Teamwork Collection brings together Jira, Confluence, Loom, and Rovo, and it’s always been about helping teams work together more easily. The new features across Teamwork Collection target the parts that were holding AI back: the quality of the context behind it and the collaboration around it. Connect your context, keep it current, and put it to work, with a person in the loop while agents do the grunt work.
Tell us how these land and what you want to see us ship next.
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.
