AI news · October 8, 2026
Atlassian connects GPT-6 agents to Rovo's Teamwork Graph
Atlassian says Rovo can use the Teamwork Graph to assess launch readiness, blockers, and missed milestones. More than 3,000 Atlassian developers reportedly use Codex in daily development work.
- Atlassian developers using Codex
- 3,000+
- named work sources: Jira, Confluence, discussions
- 3
Atlassian and OpenAI announced a partnership on October 6 that puts GPT-6 family models into Atlassian's Rovo agents. The companies describe Rovo using the Teamwork Graph to combine information from Jira, Confluence, and team discussions, such as checking launch readiness, identifying blockers, and finding missed milestones. OpenAI says more than 3,000 Atlassian developers use Codex across terminal work, IDE work, and code review.
The value is in the connected work data, not in a model name by itself. An agent can only give a useful project answer when the underlying issues, documents, and decisions are current and accessible. Teams should start with a read-only project question and check every cited source before allowing updates. The partnership also raises a practical data question: the time saved by one connected answer must be measured against permission setup, stale records, and review time.
What you can do with it
Try one read-only Rovo question about a live project, such as launch blockers or missing owners. Open each supporting Jira or Confluence record, note stale data and wrong joins, and compare the result with a manual project review. Do not grant update rights until source accuracy is clear.
Our take
The partnership targets a real business problem: project facts are spread across tools. The product will win only if the graph is current and the agent shows its source trail clearly. More model capacity cannot fix missing updates, bad permissions, or a team that does not maintain its records.
Links Partnership
Source: OpenAI and Atlassian ↗ — Made With Models writes the brief; the reporting is theirs.