Last updated on 21st February 2026

AI Code Review

DeployHQ's AI Code Review feature analyzes your code changes before deployment and provides actionable insights on security, performance, code quality, and more. Reviews are generated automatically using AI, so you can catch potential issues before they reach production.

Requesting a Code Review

When creating a new deployment, you'll see a Code Review section below the start and end revisions. Click Request AI Code Review to generate a review of the changes between the selected revisions.

New Deployment - Code Review Section

The review is generated as a preview deployment, so no files are uploaded to your server. Once the review is complete, you'll be redirected to the report page automatically.

Reading the Review Report

The report includes an overall score out of 100, a summary of findings, and detailed recommendations grouped by category.

AI Code Review Report

Overall Score

The score at the top of the report gives you a quick indication of the overall quality of the changes. Higher scores indicate fewer or less severe issues.

Deploy Recommendation

Below the deployment details, the report shows a deploy recommendation banner. The AI analyzes all findings and provides a clear verdict:

  • Safe to Deploy (green) -- No critical or high severity security issues, data loss risks, or breaking changes were found.
  • Review Before Deploying (red) -- The review identified issues that should be addressed before deployment, with a brief explanation of the concerns.

Deploy Recommendation Banner

When critical issues are found, the banner also displays the count of critical findings.

Findings

Each finding includes:

  • Severity -- Critical, High, Medium, Low, or Info
  • Category -- Security, Quality, Performance, Testing, Documentation, Maintainability, Accessibility, Browser Compatibility, SEO, or WordPress
  • Description -- An explanation of what was found
  • Recommendation -- A suggested action to resolve or improve the issue

Critical and high severity findings are highlighted to draw attention to the most important issues.

Feedback

After reviewing the report, you can rate it as helpful or not helpful using the feedback bar at the bottom. This helps improve future reviews.

Limits

  • Each deployment can have up to 3 code reviews. If a review fails, you can retry it.
  • Your account may have a daily review limit depending on your plan. The remaining reviews for today are shown when the limit is approaching.

Retrying a Failed Review

If a review fails due to a temporary issue (such as a network timeout), click Try Again on the error page to delete the failed review and start a new one.

Proceeding to Deploy

After reviewing the findings, click Run Deployment to proceed with deploying the changes, or Back to Deployment to return to the deployment form and adjust your settings.