Atlassian has announced the latest addition to its Rovo MCP ecosystem with the launch of a new Model Context Protocol (MCP) connector for ChatGPT.
The new connector gives ChatGPT users access to Jira and Confluence data, allowing them to summarize tasks, create issues, and automate workflows directly using the chatbot.
"Whether you’re pinpointing best practices from a recently completed campaign, surfacing trends in recent P1 incidents, or identifying which upcoming initiatives have the greatest potential for success, the Atlassian Rovo MCP Server and ChatGPT make it easy to connect the dots across your organization’s knowledge," the firm in a blog post.
Users can now summarize Jira work items or Confluence pages instantly within ChatGPT and go beyond a basic summary to make use of the AI tool’s reasoning and research capabilities.
Similarly, they can create Jira issues directly within ChatGPT, automate multi-step actions such as generating or updating issues in bulk, and enrich Jira content with context pulled from multiple sources accessible within ChatGPT.
"If you’re in engineering and ops, you can kick off your day with a standup summary that writes itself. Just started a marketing project? Agents can monitor tasks for deliverables, summarize their status, and suggest next actions based on what’s on track, at risk, or blocked," said Atlassian.
"And for our friends in customer support, lean on agents to triage new tickets, suggest responses based on your knowledge base and past cases, and even auto-create Jira issues for bugs."
Atlassian enhances Rovo MCP Server
The company has also enhanced the Atlassian Rovo MCP Server to allow Atlassian tools to be connected to third-party MCP clients securely. It uses OAuth authentication and respects all existing permission controls, the firm explained, so that data stays protected wherever it’s accessed.
Atlassian has now added audit logs that provide full visibility into how data is accessed and which actions are taken, along with tool invocations to track when and how MCP tools are used.
Elsewhere, a new ‘supported domains allowlist’ feature gives admins control over which clients can connect to the Rovo MCP Server, allowing users to explicitly approve MCP clients of their choice and connect tools beyond the firm's officially supported connectors.
"This combination of guardrails and flexibility ensures teams can unlock powerful AI integrations without ever compromising on security, transparency, or governance," said the company.
Atlassian Rovo Model Context Protocol (MCP) Server was released in beta earlier this year. Since then, the company has added more than 20 MCP connectors, partnering with Figma, HubSpot, and Lovable.
"MCP helps us deliver on our promises of open connectivity, accelerated workflows, context-rich AI, customer choice, and a vibrant builder ecosystem," said the company.
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
MORE FROM ITPRO
-
Technical standards bodies hope to deliver AI success with ethical development practicesNews The ISO, IEC, and ITU are working together to develop standards that can support the development and deployment of trustworthy AI systems
-
AWS CISO Amy Herzog thinks AI agents will be a ‘boon’ for cyber professionalsNews AWS CISO Amy Herzog thinks AI agents will be a ‘boon’ for cyber professionals, and the company has already unlocked significant benefits from the technology internally.
-
AWS says ‘frontier agents’ are here – and they’re going to transform software developmentNews A new class of AI agents promises days of autonomous work and added safety checks
-
Breaking boundaries: Empowering channel partners to unite DevOps and MLOps for a stronger software supply chainIndustry Insights Unifying DevOps and MLOps speeds delivery, strengthens governance, and improves software supply chain efficiency
-
Google CEO Sundar Pichai thinks software development is 'exciting again' thanks to vibe coding — but developers might disagreeNews Google CEO Sundar Pichai claims software development has become “exciting again” since the rise of vibe coding, but some devs are still on the fence about using AI to code.
-
Open source AI models are cheaper than closed source competitors and perform on par, so why aren’t enterprises flocking to them?
Analysis Open source AI models often perform on-par with closed source options and could save enterprises billions in cost savings, new research suggests, yet uptake remains limited.
-
‘Slopsquatting’ is a new risk for vibe coding developers – but it can be solved by focusing on the fundamentalsNews Malicious packages in public code repositories can be given a sheen of authenticity via AI tools
-
Microsoft’s Windows chief wants to turn the operating system into an ‘agentic OS' – users just want reliability and better performanceNews While Microsoft touts an AI-powered future for Windows, users want the tech giant to get back to basics
-
Google Brain founder Andrew Ng thinks everyone should learn programming with ‘vibe coding’ tools – industry experts say that’s probably a bad ideaNews Vibe coding might help lower the barrier to entry for non-technical individuals, but users risk skipping vital learning curves, experts warn.
-
European software spending is set to surge in 2026 – here's whyNews Enterprises are approaching the “trough of disillusionment” with AI, but it’s not stopping them from spending
