A senior design review on every pull request. Install the App once and every UI change gets scored 0 to 100, with one-click fixes posted inline, in about a minute.
Every comment includes the issue, its severity, the affected UI area, and a concrete fix, directly inside the pull request where engineers already work.
The hero has "Get Started" and "Learn More" both styled as filled white buttons at the same size. Two primary-weight buttons in one row don’t make a hierarchy.
The +12% indicator uses an inline color, hardcoding a value that should live in tokens.
All six feature cards share identical weight, color, and spacing. Nothing draws the eye to any single card.
Install once, then every pull request is reviewed. No CI config, no API keys, no workflow files.
One-click install from the GitHub Marketplace. No CI config, no API keys, and no workflow files to set up or maintain.
Push your UI changes to any branch and open a pull request the way you already do. Rams picks it up automatically.
Rams reads the changed files and posts inline comments plus a top-level summary, usually within about a minute.
Accept each suggestion with one click, or dismiss the ones you disagree with. Then merge with confidence and ship clean UI.
Not linter rules. Review judgments, versioned and public in the rules. New ones ship every week. Every repo Rams reviews gets sharper without changing a thing.
Semantic HTML, keyboard nav, focus, contrast
Token bypasses, contrast ratios, semantic use
Hierarchy, scale, line height, truncation
Scale adherence, gap usage, concentric radii
Composition, reuse, token leaks, prop APIs
Loading, empty, and error states, form patterns
Duration, easing, reduced-motion, layout shift
AI antipatterns: magic numbers, slop gradients
Everything a review gives your team, and everything it never asks of your setup.
Every pull request from every contributor gets reviewed, humans and agents alike. Nobody else has to install anything.
Findings arrive as inline suggestions with severity and file:line. Commit each fix straight from the pull request.
0–100, severity-weighted. Any critical issue caps it at 59, so 60+ always means zero criticals.
Push fixes and Rams re-reviews, checking every finding by name. The score climbs as the work lands.
Every review lands in the dashboard: scores across PRs and repos over time, so design quality becomes a trend, not a debate.
One-click install from GitHub. No workflow files, no API keys, no pipeline changes.
Same engine, different moments. The MCP reviews code inside your agent before a commit. The App is the backstop that reviews everything that reaches a pull request. Most teams run both on one plan.
Whether it blocks merges, what it reviews, what it costs, and what leaves GitHub.
No. Reviews post as pull request comments with a score and inline fix suggestions. The merge stays yours. Most teams treat a critical finding the way they treat a failing human review: fix it before merging, but nothing is forced.
Every pull request review draws from the same monthly pool as MCP reviews: free 30, Solo 50, Basic 300, Pro 750 per month. One plan, no separate charges per surface.
The UI files in the pull request diff: components, styles, and markup. Backend-only pull requests pass through untouched and don’t count against your quota.
The changed files in the pull request are sent to the Rams worker over HTTPS, reviewed, and returned as comments. File contents are not used to train models.
Install Rams on a repo and get your first design quality review on the next pull request.
Frameworks