AIToday
Large Language ModelsAI Coding AssistantsPublickeyPublished: Jul 23, 2026, 01:00 JST

Atlassian adds AI requirement auto-creation to Jira, routes tasks to Claude, Copilot

Atlassian adds AI requirement auto-creation to Jira, routes tasks to Claude, Copilot

3 Key Points

  1. What happened

    Atlassian announced new AI features for Jira, including Jira Planner (which automatically generates requirements from code and Confluence context), task identification, and assignment of work items to humans or AI agents like Claude Code, GitHub Copilot, Cursor, and a new built-in Jira Coding Agent.

  2. Why it matters

    Development teams can now keep task context in Jira while flexibly routing work to different AI coding tools, reducing manual requirement writing and task breakdown. The system tracks AI agent progress and can automate routine work like bug fixes, vulnerability repairs, test generation, and documentation updates.

  3. What to watch

    Jira Coding Agent now supports Claude Code, Cursor, and GitHub Copilot, with OpenAI Codex support coming soon. Engineers receive notifications when pull requests are ready.

Not sure about something? Ask the AI

Summaries like this, in your inbox every morning.

Context & Analysis

Atlassian's update addresses a core friction point in AI-assisted development: keeping context synchronized across multiple tools. By embedding requirement generation directly into Jira and anchoring all task assignments to that single source of truth, teams avoid the context loss that occurs when jumping between Jira, GitHub, IDE-based tools (Copilot, Cursor), and external LLMs. The integration with Claude Code, GitHub Copilot, and Cursor reflects the current fragmentation of AI coding tools—developers often prefer different agents for different tasks—and Jira's new design allows that flexibility without breaking the workflow.

The system also formalizes the role of AI agents as assignees, not just suggestions. This suggests Atlassian sees future development as genuinely mixed human-AI teams, where task routing becomes a first-class concern. The addition of automated routine work (tests, docs, vulnerability patches) via the automation rules builder further reduces friction, since these tasks can now run without explicit assignment. The notification of engineers when pull requests are ready completes the loop—AI does the heavy lifting, humans gate the final decision.

FAQ
Which AI agents can be assigned tasks in Jira?
Claude Code, Cursor, GitHub Copilot, and the built-in Jira Coding Agent are currently supported, with OpenAI Codex support coming soon.
How does Jira Planner generate requirements?
Jira Planner retrieves context from the existing codebase, Jira history, and Confluence, then uses that information to automatically create structured requirement documents.
What types of work can be automated with the new system?
Bug fixes, vulnerability repairs, test generation, and documentation updates can be automatically routed to AI agents using Jira's automation rules builder.

Get the latest Large Language Models news every morning

For example, today's edition would include:

  • Abeam and Notion target enterprise knowledge for AI agentsITmedia AI+ · 3h ago
  • Zscaler unveils Agentic SOC with AI agentsITmedia AI+ · 3h ago
  • Generative Partners launches "AX BPO" for work AI alone can't finishITmedia AI+ · 3h ago

AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. Q&As are published on this page for other readers too.

Related Articles

Next articleOpenAI model exploited zero-day vulnerabilities; calls for no-fault AI liability