Introduction to Solution Reports
When you submit a solution on Frontend Mentor, our system automatically analyzes your code and generates detailed feedback reports. These reports help you identify areas for improvement and reinforce what you're doing well.
Table of contents
What solution reports are
Every time you submit a solution, we automatically review your code and tell you what you did well and where you can improve. There are two parts to this: automated checks that run on every submission, and a deeper AI code review.
Automated checks
Automated checks run on every submission, for everyone. We use established linting tools to catch concrete, rule-based issues across four areas:
- Accessibility, using axe-core, for issues like missing alt text, form labels, and heading order
- HTML, using html-validate, for structure and semantic markup
- CSS, using stylelint, for maintainability and common pitfalls
- JavaScript, using eslint, for code quality and likely bugs
Each issue is grouped by area and flagged by severity: error, warning, or info. Automated checks don't include a score. They point you to specific lines worth fixing. Each area has its own guide, linked at the bottom of this page.
The AI code review
The AI code review goes further than rule-based checks. It reviews your whole solution the way a thoughtful reviewer would, then scores it across seven dimensions grouped into three categories:
- Code Quality: Best Practices, Architecture, File Organization, and Testing
- Design & UX: Accessibility
- Documentation: README Quality and Commit Quality
You get an overall score out of 10, plus a score and tailored feedback for each dimension. Findings are grouped into strengths, suggestions, and things that need attention.
Scores and rating bands
Each score, from the overall down to each dimension, comes with a rating band so you can read it at a glance: Needs Work, Fair, Solid, Great, and Exceptional.
Difficulty calibration
The review is calibrated to the challenge difficulty, so the bar rises as you take on harder projects. Some dimensions only apply once a challenge reaches a certain level: File Organization starts at Junior, Architecture at Intermediate, and Testing at Advanced. A dimension that doesn't apply to a challenge shows as locked, and it doesn't count toward your overall score.
Who gets the AI code review
- Pro members get an AI code review on every submission.
- Free members get one AI code review a month. After that, your submissions get the automated checks until your credit refreshes the following month.
When reports are generated
Reports run automatically after you submit:
- You submit your solution with your repository URL.
- We pull your code from the repository.
- The checks run, along with the AI code review if you have a credit.
- Your report appears on your solution page.
It usually takes a few minutes.
Report privacy
Your reports are private. Only you can see the detailed feedback, so you can review it and work on weak spots without worrying about anyone else seeing them.
How to access your reports
- Open your submitted solution.
- Find the report section on the solution page.
You can also click the "Improve Solution" step on the challenge page to see the report for that project.
Making the most of your reports
Look across every area. Strong frontend work spans accessibility, semantic HTML, maintainable CSS, solid JavaScript, and good documentation habits, so reviewing all of your feedback helps you grow as a well-rounded developer.
Finding things to fix is normal. Even experienced developers get feedback on their code, so treat each issue as something to learn from rather than a mark against you.
Watch for patterns across challenges too. If one area is consistently lower than the rest, that's a clear signal for where to focus next.
Related guides
Automated checks:
AI code review:
And when you're ready to act on what you find: Acting on Report Feedback.