As teams adopt AI coding agents, they're writing code faster and in greater volume. That puts pressure on the rest of the software lifecycle, and code review especially has become one of the biggest constraints on getting software into production.
Developers now face more code to review and bigger changes. According to our Q3 State of AI Impact in Engineering Report, teams that rely heavily on AI to write code now wait more than twice as long for a human review on a pull request (71 minutes, up from 32). And median pull request size has more than doubled in the past year from 46 to 102 lines of code. Human review alone is struggling to keep up, and that's the problem Code Reviewer is built to solve.
Code Reviewer is available now as part of your paid Jira or Teamwork Collection subscription
Code Reviewer is an AI agent that reviews pull requests in Bitbucket Cloud and GitHub (with GitLab coming soon). Because it's deeply connected to Jira and the Teamwork Graph, each review is grounded in your context. That means every change gets a fast and useful first review, and your teammates can focus on the feedback that needs human judgment.
Code Reviewer leaves inline comments to flag issues and suggest fixes authors can apply right in the pull request. It catches bugs, logic errors, security risks, and maintainability issues. It checks pull requests against acceptance criteria in linked Jira work items. And for large changes, it also adds a summary to help reviewers get oriented.
Every comment should be worth a reviewer's time. That’s why a built-in quality filter removes low-value comments, so the feedback that reaches your team is signal, not noise.
We've seen the results on our own code. Since rolling out Code Reviewer internally, median pull request cycle time has dropped by 31%. And 39% of Code Reviewer's comments led directly to a code change in the next commit.
Bring your coding standards into every review (beta)
With Standards, a new feature available in beta you can document your organization's coding standards and assign them to specific repositories. Code Reviewer then checks each pull request against that repository's standards, so your requirements are applied consistently, no matter who's reviewing. Sign up for the beta here.
Get a review on every pull request, automatically
Code Reviewer is on by default for eligible sites, in manual mode. It only reviews a pull request when the author asks. To get the most out of Code Reviewer, turn on automatic review, so every new pull request gets an AI first pass as soon as it opens, without anyone having to ask. Watch the demo or read our docs for how to set it up.
Your code stays yours. You stay in control
Trust and safety are built into how Code Reviewer works. Each review runs in a temporary, isolated environment that is deleted when the review finishes. Your code is not stored afterwards and is never used to train models.
Admins decide where automatic reviews run, which workspaces, which repositories, and can switch any of them back to manual or turn Code Reviewer off.
Get a review on every pull request, automatically
Code Reviewer is on by default for eligible sites, in manual mode. It only reviews a pull request when the author asks. To get the most out of Code Reviewer, turn on automatic review, so every new pull request gets an AI first pass as soon as it opens, without anyone having to ask.
How to turn on automatic review
Before you start: Code Reviewer needs your code connected to Jira. For GitHub, install the GitHub for Atlassian app from the Atlassian Marketplace. For Bitbucket Cloud, connect your Bitbucket workspace to your Jira site.
Bitbucket Cloud (workspace or repository admin)
- Open the settings: In Bitbucket, go to Workspace settings > AI, and select Code Reviewer.
- Choose when Code Reviewer runs: Change the trigger schedule from “Manual triggers only” to “Every new pull request”. For the most coverage, choose Every commit so that each new commit to a pull request gets reviewed again.
- Make it the workspace default: Select “Set as default”. Every repository in the workspace then uses this setting.
- Fine-tune a repository (optional): Go to Repository settings > Features > Code review to change the setting for one repository.
GitHub (Jira site admin)
For GitHub repositories, Code Reviewer is managed from Jira.
- Open the Code Reviewer settings: In Jira, you can access settings in two places.
- For Jira site admins to control site level access: Go to Jira site settings > Developer AI > Reviewer settings and select GitHub.
- For source code management admins: Go to Development tab > select Connection tab on the page > find Code Reviewer and select settings
- Choose your organization: Select the GitHub organization, then select Manage settings for all repositories.
- Turn on automatic review: Set the trigger to “When PR is created”, then select Apply to all repositories in this organization.
- Fine-tune a repository (optional): Select any repository in the list to change its settings. You'll find more options there.
Ready to start?
If you are an admin: turn on automatic review today using the steps above, so every new pull request gets a review from the moment it is opened.
If you are not an admin: share this post with your Bitbucket workspace admin or Jira site admin and ask them to turn on automatic review for your team.
Already using Code Reviewer through Rovo Dev? Your access continues.
Questions? Visit the Code Reviewer documentation. Feedback? Share your thoughts through our in-product feedback collector.
