Cursor Origin
Reviewing Agent Code on Cursor Origin
Reviewing agent code means judging changes you didn't write. Today Cursor gives you Bugbot for pull-request review and Auto-review to gate risky changes, on top of Graphite's stacked review. Origin reportedly adds mobile review of agent diffs and inline comments, but those specifics aren't yet confirmed by Cursor.
On this page
How do you review code you didn't write?
When an agent writes the change, the writing stops being the hard part. Reading it becomes the job. That is a different skill from authoring: you are reconstructing intent from a diff, checking it against what you actually asked for, and deciding whether you trust it enough to merge.
The trap is volume. Agents produce changes faster than you can carefully read them, and the temptation is to approve on vibes. The whole point of a review layer built for agents is to keep human attention on the changes that need it and take the rest off your plate.
What review tools does Cursor give you today?
You don't have to wait for Origin to put review automation in front of agent output. These ship now and work on GitHub and GitLab.
Reviews pull requests and flags likely bugs before a human reads them. Cursor reports it getting faster and cheaper over time.
Governs agent autonomy by gating risky changes for a closer look, so not everything lands on trust.
Read the agent's diff in Cursor as it works, before anything reaches a pull request.
Graphite, the team behind Origin, also ships an AI reviewer (Diamond) and a stack-aware merge queue. Those are Graphite products today, and they hint at the shape of Origin's review layer, but they are not the same thing as Origin.
What does Origin's review layer reportedly add?
Coverage of the Compile demo described a review experience aimed at agent work specifically, including from your phone. The pieces below are the ones reported. Cursor's cloud agents do already have web and mobile control, so the direction is real even where the Origin specifics aren't confirmed.
- Reviewing agent-generated diffs and leaving inline comments, as you would on a pull request.
- Looking at screenshots of an agent's visual output, useful when the change is UI.
- Starting, unblocking and redirecting agent tasks remotely, including agents running on your local machine.
BugbotCursor's automated PR reviewer that posts inline findings and can push fix commits from isolated VMs., Auto-review and in-editor agent review are confirmed Cursor features. The mobile review of agent diffs, inline comments and screenshot review described for Origin come from press coverage of the demo, not a Cursor source. Verify them on cursor.com/origin at launch.
What should a human still do at review?
Interactive diagram. Tab through its regions; each focused region shows its detail in the panel below.
Bots take the first pass and the easy gate. The human spends attention at the merge.
A bot can tell you a change is probably safe. It can't tell you it's the change you wanted. Reviewing agent code is mostly checking intent against outcome, and that stays a human job no matter how good the review layer gets.
Frequently asked questions
Can Bugbot review my agent's code today?
Yes. Bugbot reviews pull requests and flags likely bugs on GitHub and GitLab now, independent of Origin. It's the most direct way to put an automated first pass in front of agent-generated changes while Origin is still pre-launch.
Does Origin let me review code from my phone?
It reportedly does, based on coverage of the Compile demo, which described reviewing agent diffs, leaving inline comments and viewing screenshots of visual output on mobile. Cursor's cloud agents already have mobile control, but the Origin-specific review features aren't confirmed on a Cursor source yet.
What is the difference between Bugbot and Auto-review?
Bugbot reviews changes and points out likely bugs. Auto-review is about governance: it decides which agent changes are risky enough to hold for a human instead of letting them proceed. One finds problems; the other controls how much autonomy an agent gets.
Sources & last verified
- Cursor — Bugbot
- Cursor — Governing agent autonomy with Auto-review
- Graphite — AI code review (Diamond)
- Cursor — Origin (waitlist)
Cursor ships frequently. Facts verified against primary sources on June 26, 2026.