Front-End Checklist is an open-source quality assurance system designed to help both human developers and AI agents systematically review frontend code and implementations. Maintained primarily by thedaviddias with contributions tracked across multiple developers, the project provides a comprehensive corpus of 385 English rules organized across 11 active categories covering HTML, CSS, JavaScript, performance, accessibility, SEO, security, images, testing, privacy, and internationalization. The repository is written in MDX and serves as the source for a web-based interface at frontendchecklist.io as well as an MCP-compatible server for agent-based workflows.
The project addresses a practical gap in frontend development by converting abstract best practices into actionable, prioritized review workflows. Rules are classified by severity using a four-tier priority system: critical for site-breaking or compliance-sensitive issues, high for major user experience or accessibility impacts, medium for strong best practices in normal quality review, and low for situational improvements. This structured approach allows developers and teams to focus remediation efforts on the most impactful issues first. The checklist spans multiple domains including responsive design, code quality, website launch readiness, performance optimization, and usability.
Users can engage with Front-End Checklist through three primary workflows. The web interface at frontendchecklist.io provides interactive browsing and filtering of all rules with curated checklists for specific scenarios. The MCP server at mcp.frontendchecklist.io exposes 11 tools that enable AI agents to review code, audit live URLs, search rules by keyword or category, and fetch specific rules with remediation guidance. The README itself serves as a single-document reference for developers who prefer working directly in text form. The project also maintains a companion resource called UX Patterns for Devs to help developers select appropriate UI patterns before using Front-End Checklist to verify implementation quality.
The repository supports local development through pnpm-based tooling that includes validation of rule structure, scoring of the rule corpus, and generation of derived artifacts like skills and README updates. Contributors can validate their work against the established rule structure before submission. GitGenius activity tracking shows a median issue and pull request response latency of 62.3 hours with a mean of 224.3 hours, indicating active but measured maintenance. The most frequently applied labels are stale and wontfix, suggesting the project manages scope carefully. The repository shares contributors with major projects including microsoft/vscode, microsoft/typescript, and rust-lang/rust, indicating it draws from and contributes to the broader developer tools ecosystem.
The MCP integration represents a significant feature for modern AI-assisted development workflows. Agents can review pasted code against the checklist, audit public URLs for compliance with frontend best practices, search for rules by specific concerns like accessibility or performance, and retrieve complete remediation guidance with code examples. The project provides explicit guidance for agent usage, recommending that users name the MCP server in prompts to ensure discovery and suggesting entry points like code review for pasted components or URL audits for live sites. Skills-based workflows allow reusable audit patterns and focused rule-specific guidance for tools that support skill installation, enabling both broad frontend audits and narrow, concern-specific reviews.
