Skip to main content
Tusk Review is an AI code reviewer that runs on PRs to find bugs and code quality issues. The @use-tusk bot will leave inline comments for any issues and include a fix prompt, so you can resolve issues quickly before merge.

​
How It Works

1

Create or update a PR

Tusk Review runs when a PR is created or when new commits are pushed. Trigger Tusk Review at any time by commenting @use-tusk review in the PR.
2

Tusk scans for bugs and issues

AI evaluates changed files for bugs, security issues, performance/reliability problems, and maintainability concerns. When Tusk finds an issue, @use-tusk posts an inline PR comment with the risk and suggested fix.
3

Use AI prompt to investigate

Each finding includes an AI prompt you can copy and paste into your coding agent to investigate and fix the issue.
4

Apply fixes and push changes

Update your code based on the review comments, then push commits to the same PR branch.
5

Tusk re-reviews automatically

Tusk re-runs on the latest commit and updates feedback, helping you verify that previous findings are resolved before merge.

​
Trigger a review on demand

Beyond the automatic run on PR creation and new commits, you can request a review any time by commenting on a PR:
The PR author must have an active Tusk seat for the review to run.

​
Setup

This guide assumes you have already connected your repo. Tusk Review is stack-agnostic and does not require a test execution environment.
1

Choose a repo

In the Repos page, click Code Review for the repository you want to configure.
2

Enable code review

Enable the toggle to turn on Tusk Review for all PR authors with an active Tusk seat. Toggle other repo-level settings as needed.
3

Optional: Configure individual preferences

To change individual preferences for how Tusk Review runs, go to Settings > GitHub/GitLab Seats, click Edit preferences, and select the relevant checkboxes under Code Review Configuration.