Introduction
Vetoo is an agentic AI code reviewer for GitHub and GitLab.
Vetoo reviews every request your team opens on GitHub and GitLab. A request is a pull request on GitHub, or a merge request on GitLab. Vetoo reads the change, then posts findings with concrete suggestions.
Every finding carries a severity. The severity tells you what to fix now and what can wait. To question a finding, reply in the thread and mention @vetoo-review. Vetoo answers in the same thread.
What a review looks for
A review looks for flaws in logic, security holes, missing error handling, broken tests, and fragile design. Vetoo raises each one as a finding. To find them, Vetoo reads each change against the whole system, not line by line. This wider view also catches findings that reach into the systems your code connects to.
Vetoo reviews any language, from the most widely used to the most niche.
Panel of agents over a single model
Any single model has blind spots. Vetoo therefore runs a panel of agents over the same change, and the agents run on different models. Each agent raises findings that the other agents miss. A reconciliation step then merges those findings into one ranked verdict. The verdict therefore reflects every model in the panel.
Start here
Quickstart
Connect a repository and get your first review in a few minutes.
Connect GitHub
Install the GitHub App and choose the repositories Vetoo can read.
Connect GitLab
Connect gitlab.com or a self-managed instance.
How a review happens
Follow each step, from the moment a request opens to the final verdict.