mrge’s AI reviewer helps software teams ship code faster by providing immediate, actionable feedback on pull requests.

It learns your repository context, coding patterns, and adapts based on your feedback to deliver more accurate reviews.

​
Key features

  1. Automated code reviews

    • Scans pull requests for logic errors, style inconsistencies, and potential vulnerabilities.
    • Leaves structured comments and suggestions directly in the PR.

  2. Codebase awareness

    • Detects libraries and frameworks in your repo (including their versions).
    • References relevant code patterns or utilities found in your codebase when suggesting improvements.

  3. Custom rules

    • Enforce specialized guidelines, such as domain-specific standards or internal best practices.
    • Automatically flags violations based on your configured rules.

  4. Adaptive feedback

    • Learns from your team’s past merges, comments, and preferences.
    • Improves over time to reduce false positives and tailor advice to your codebase.

​
How it works

1

Enable AI review

Enable AI review for your repository in the AI review page.

2

Open a pull request

Once a new PR is created, mrge scans the changed files and runs its analysis.

3

Automated review

The AI leaves comments regarding potential issues or recommended improvements.

4

Custom rules check

If you have custom rules in place, the AI will also flag any violations based on those criteria.

5

Feedback and collaboration

If a comment needs clarification or correction, you can reply to the AI comment to refine its behavior for future reviews.

6

Merge or iterate

Address, ignore, or further discuss the flagged items. When ready, merge the PR.

Currently, mrge analyzes the code once per pull request creation. If you commit additional changes, the AI will not automatically re-check those (this feature is planned for a future release).

​
Next steps

Was this page helpful?