Skip to content

Guide · Open source

How to use Front-End Checklist: browse, run audits and use with AI agents

Updated · written with AI from the project's own documentation and checked against our catalogue data

Front-End Checklist is an MIT-licensed collection of 386 English rules covering HTML, CSS, JavaScript, performance, accessibility, SEO, security, images, testing, privacy and internationalization. Each rule describes a practice to check and links to a rule page with explanations and verification steps. This guide shows you the three ways to use it and how to run a focused review with it.

Front-End Checklistthedaviddias/Front-End-Checklist · ★ 74.6k · MIT

Quick start

  1. 1Pick your entry point. Choose the website for browsing, the MCP server for AI agents, or the README for a single checklist.
  2. 2Choose a category. Select the category that matches your audit, such as Accessibility, SEO or Performance.
  3. 3Check Critical and High items first. Priority markers show which rules are site-breaking, security-sensitive or high impact.
  4. 4Open the rule page. Each rule links to explanations, remediation guidance and verification steps.
  5. 5Run an MCP audit (optional). Connect an MCP-capable agent to mcp.frontendchecklist.io and ask it to use the Front-End Checklist MCP on your component or public URL.
  6. 6Record your findings. Tick off the items that apply and note which ones still need fixing.

Choose how you want to use it

You can use Front-End Checklist in three ways. Browse the rules on the website at frontendchecklist.io, connect an MCP-capable AI agent to the hosted server, or work through the checkboxes in the README directly.

The website is best for browsing and filtering. The MCP server is for agent workflows. The README is useful when you want the whole checklist in one place.

  • Website: frontendchecklist.io, with rules at frontendchecklist.io/rules and curated lists at frontendchecklist.io/checklists
  • MCP server: mcp.frontendchecklist.io, documented at frontendchecklist.io/mcp
  • README: the checklist grouped by category with checkboxes you can tick off

Work through the checklist by category

The checklist is split into 11 categories: HTML (25 rules), CSS (32), JavaScript (26), Performance (44), Accessibility (95), SEO (94), Security (22), Images (25), Testing (13), Privacy (5) and Internationalization (5). Start with the categories that match your project or audit.

Each item has a priority marker. Critical items are site-breaking, compliance-sensitive or security-sensitive and should be fixed first. High items have major impact on user experience, accessibility, performance or discoverability. Medium items are strong best practices for normal quality review. Low items are situational.

  • Tick items that apply to your project, audit or pull request
  • Open the linked rule page for explanations, examples, verification steps and AI prompts
  • Fix Critical items before Medium or Low ones

Use it with an MCP-capable AI agent

Connect an MCP-capable agent to the hosted endpoint at mcp.frontendchecklist.io. The server exposes 11 tools. The README recommends starting with review_code for pasted HTML, CSS, JavaScript, React or Next.js code, and audit_url for public https:// pages.

Some clients discover installed MCP tools lazily, so name the server in your prompt. The README gives this example: Use the Front-End Checklist MCP to review this React component and report the highest-confidence findings first.

  • review_code: pasted HTML, CSS, JavaScript, React or Next.js code
  • search_rules: look up rules by keyword, category or priority before making recommendations
  • get_workflow or get_checklist_rules: launch, accessibility, SEO, security and performance audits
  • audit_url: public https:// pages
Use the Front-End Checklist MCP to audit https://example.com for accessibility, performance, and SEO issues.

Install the agent skills package

If your tool supports skills, you can install the Front-End Checklist skills package. The global skill is an audit entry point across the full corpus. Focused skills cover one concern, such as the https skill for security review.

The README lists two install commands. The first installs the whole package and the second installs only the https skill.

  • Global audit entry point: skills/frontend-checklist-global/SKILL.md
  • Focused example: skills/https/SKILL.md
npx skills add frontendchecklist/skills npx skills add frontendchecklist/skills --skill https

Run a focused pre-launch or pull request review

Pick one concern for your first pass, such as accessibility or SEO, rather than the whole list at once. Check the rules in that category, then open the rule pages for anything unclear.

For pull requests or shared audits, the README checklist works well as a common reference. The README describes this as a review workflow. It does not run automatically against your code.

  • Start with Critical and High items in the category you are reviewing
  • Confirm each finding using the verification steps on the rule page
  • Record which items you have checked in your review notes

Contribute to the checklist

If you want to change the rules, the README lists local development scripts. Run pnpm install to install dependencies, pnpm dev for local development, and pnpm validate:rule-structure to validate rule structure. Other scripts are pnpm score:rules, pnpm generate:skills and pnpm generate:readme.

pnpm install pnpm dev pnpm validate:rule-structure

What people use it for

  • Pre-launch review of a website or web app
  • Shared checklist for pull request or accessibility audits
  • Grounding findings from an MCP-capable AI coding agent in a rule set
  • Students learning what good modern front-end practice looks like

Frequently asked questions

How do I install Front-End Checklist?

The project is used as a web app and the README lists no prerequisites. For agent skills, run npx skills add frontendchecklist/skills. You can also browse the rules directly at frontendchecklist.io.

Is Front-End Checklist free?

Yes. It is released under the MIT licence, as shown in the project badge and metadata.

Does Front-End Checklist work as a CI linter?

No. The README describes a review workflow rather than a build-time tool, so it is not designed to run automatically in CI.

Can I use it with any AI agent?

It is for AI agents that support MCP. Connect them to the hosted server at mcp.frontendchecklist.io, or install the skills package in tools that support skills.

How does it compare to Google Lighthouse, axe-core or WebHint?

The source lists these as comparable projects but does not compare them in detail. Front-End Checklist is a set of rules with explanations and verification steps for review, not an automated scanner.

Is it suitable for backend or non-web projects?

No. The scope is front-end only, so backend or non-web software needs a different checklist.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

Front-End Checklist suits developers doing pre-launch or pull request reviews, teams wanting a shared audit checklist, and people using MCP-capable AI agents who want findings grounded in a rule set. If you need an automated linter that runs in CI, or a step-by-step beginner tutorial, pick a different tool.

Install options & downloads