About Accessibility Hub
Accessibility Hub is an MCP server for web developers that orchestrates axe-core, Pa11y, and Lighthouse accessibility checks. It adds WCAG context, affected-user details, and remediation guidance to findings.
Key features
- 01Run axe-core, Pa11y, and Lighthouse checks
- 02Combine results from parallel analyses
- 03Check WCAG 2.1 and APCA color contrast
- 04Add WCAG context and affected-user details
- 05Provide remediation priorities and steps
Use cases
- 01Audit a page and obtain a WCAG compliance report
- 02Check accessibility score thresholds before deployment
- 03Find contrast issues and review suggested fixes
MCP support
The MCP server exposes web accessibility analysis tools, prompts, and reference resources such as WCAG criteria to clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEAccesibilityHub
MCP Server for orchestrating web accessibility tools (axe-core, Pa11y, Lighthouse). Analyze web pages, check color contrast, get Lighthouse accessibility scores, and get detailed WCAG compliance reports with enriched human context.
Quick Start
npm install -g AccesibilityHub
Add to your MCP client configuration:
{
"mcpServers": {
"AccesibilityHub": {
"command": "npx",
"args": ["-y", "AccesibilityHub"]
}
}
}
Then ask your AI assistant:
"Analyze the accessibility of https://example.com"
Available Tools
| Tool | Description | Docs |
|---|---|---|
analyze-with-axe | Analyze accessibility using axe-core | docs/tools/analyze-with-axe.md |
analyze-with-pa11y | Analyze accessibility using Pa11y | docs/tools/analyze-with-pa11y.md |
analyze-with-lighthouse | Analyze accessibility using Lighthouse (score 0-100) | docs/tools/analyze-with-lighthouse.md |
analyze-contrast | Check color contrast (WCAG 2.1 / APCA) | docs/tools/analyze-contrast.md |
analyze-mixed ⭐ | Run multiple tools in parallel and combine results | docs/tools/analyze-mixed.md |
Available Prompts
| Prompt | Description | Docs |
|---|---|---|
full-accessibility-audit | Comprehensive audit using axe-core, Pa11y, and Lighthouse | docs/prompts/full-accessibility-audit.md |
quick-accessibility-check | Fast check with critical issues summary | docs/prompts/quick-accessibility-check.md |
contrast-check | Color contrast analysis with fix suggestions | docs/prompts/contrast-check.md |
pre-deploy-check | Verify compliance with score threshold before deployment | docs/prompts/pre-deploy-check.md |
quick-wins-report | High-impact, low-effort fixes from all three tools | docs/prompts/quick-wins-report.md |
explain-wcag-criterion | Deep dive into any WCAG criterion | docs/prompts/explain-wcag-criterion.md |
lighthouse-audit | Lighthouse score-focused accessibility audit | docs/prompts/lighthouse-audit.md |
lighthouse-score-improvement | Phased plan to reach a target accessibility score | docs/prompts/lighthouse-score-improvement.md |
Available Resources
| Resource | Description | Docs |
|---|---|---|
wcag://criteria | WCAG 2.1 criteria reference data | docs/resources/wcag-criteria.md |
contrast://thresholds/* | Contrast thresholds (WCAG 2.1 / APCA) | docs/resources/contrast-thresholds.md |
lighthouse://audits | Lighthouse accessibility audit catalog with WCAG mappings | docs/resources/lighthouse-audits.md |
Enriched Human Context ✨
All accessibility issues include enriched context to help you understand and fix them:
- WCAG criterion explanation with real-world impact
- Affected users (screen-reader, keyboard-only, low-vision, etc.)
- Remediation priority and fix effort estimates
- Step-by-step suggested solutions
{
"ruleId": "image-alt",
"severity": "serious",
"humanContext": "Non-text content (WCAG 1.1.1 - Level A)...",
"affectedUsers": ["screen-reader", "low-vision"],
"priority": "critical",
"remediationEffort": "low"
}
Documentation
| Document | Description |
|---|---|
| Getting Started | Installation and MCP client configuration |
| Tools Reference | Detailed tool documentation |
| Prompts Reference | Prompt templates and usage |
| Resources Reference | Available WCAG and contrast data |
| Workflows Guide | Recommended workflows |
| Effective Prompts | Tips for better accessibility prompts |
| Interpreting Results | How to prioritize and fix issues |
| Contributing | Project structure, development, and contributions |
Requirements
- Node.js ≥ 20
- Chrome/Chromium (automatically downloaded by Puppeteer)
Support Me
If this project helps you, consider supporting its development:
License
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.
