$ loading_
Render UI artifacts, run deterministic checks, and support iterative design reviews.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "design-review" yet — see the docs or source repo.
Headlessly render this set of UI design artifacts, run deterministic lint checks, and list any layout, naming, or guideline issues you find.
A rendered result plus a reproducible issue list for step-by-step fixes.
Start a design review cycle for the current UI: render it first, then have an independent vision critic provide feedback, and save the review state for the next iteration.
A structured round of feedback and a saved review state for continued iteration.
Render the design before and after the changes, run the same lint rules on both, and summarize differences and newly introduced issues.
A comparison of version differences and a quality-change summary to help decide whether to accept the revision.
Designers can headlessly render UI artifacts and run deterministic checks before sharing a proposal. This helps catch inconsistencies or guideline issues early and reduces manual review effort.
Product managers or developers can use stateful review loops to track feedback across UI iterations. An independent vision critic helps evaluate changes across versions more consistently.
When UI revisions happen frequently, teams can apply the same deterministic lint rules across versions to detect new issues or regressions. It fits repeatable design checks where consistent results matter.
It is an MCP server for headlessly rendering UI design artifacts, running deterministic lint checks, and managing stateful design review loops with an independent vision critic.
The provided material does not include installation steps, dependencies, or configuration details. See the source repository for integration instructions.
Based on the description, it emphasizes headless rendering, deterministic checks, and trackable review loops rather than one-off subjective manual feedback. It also includes feedback from an independent vision critic.
Review design docs with multiple models and generate calibrated consensus summaries.
Review MCP servers for quality, security, scores, and improvement plans.
Search, read, and discover related design system components via MCP.
Automate code reviews with syntax checks, explanations, and improvement suggestions.
Runs dual-model code reviews and synthesizes one unified report.
Analyze Lanhu designs and Zentao bug reports to surface issues and improvements.