Magentaa11y
Pull MagentaA11y accessibility criteria into Claude Code or Cursor while reviewing UI before you ship.
Overview
MagentaA11y MCP is a Ship-phase MCP server that supplies MagentaA11y accessibility criteria to agents during UI review.
What is this MCP server?
- Exposes MagentaA11y accessibility criteria as MCP tools the agent can query on demand
- Stdio npm package magentaa11y-mcp v1.0.2 for local Claude Desktop and CLI agents
- Criteria-backed answers reduce generic accessibility hallucinations during UI work
- Fits pre-launch audits alongside manual keyboard and screen-reader checks
- GitHub source at joe-watkins/magentaa11y-mcp for self-hosting and fork audits
- Server version 1.0.2
- Transport stdio via npm identifier magentaa11y-mcp
- Single published npm package in server manifest
What problem does it solve?
Agents often give vague accessibility advice because they lack a structured, citable criteria catalog tied to your stack.
Who is it for?
Solo builders doing pre-launch accessibility review on web UIs with an MCP-enabled coding agent.
Skip if: Teams that need certified WCAG audit reports, legal compliance sign-off, or fully automated crawl-based scanning only.
What do I get? / Deliverables
Your agent can reference concrete MagentaA11y criteria while you patch UI issues before release.
- Criterion-grounded accessibility feedback in agent sessions
- Repeatable a11y checklists aligned to MagentaA11y
- Faster pre-ship UI fix loops without leaving the IDE
Recommended MCP Servers
Journey fit
Accessibility gates belong on the Ship shelf because solo builders fix a11y issues during review and pre-launch QA, not only while coding components. Review is where agents compare live markup against concrete pass/fail criteria instead of vague “make it accessible” advice.
How it compares
MCP criteria lookup server, not a browser automation scanner or a design-system skill.
Common Questions / FAQ
Who is MagentaA11y MCP for?
Indie developers and small teams who want agents to ground accessibility suggestions in MagentaA11y criteria during ship-time review.
When should I use MagentaA11y MCP?
Use it while reviewing components, landing pages, or flows before launch, or when refactoring UI and you want criterion-level checks in the agent chat.
How do I add MagentaA11y MCP to my agent?
Install the magentaa11y-mcp npm package and add a stdio MCP server entry pointing at that binary in Claude Code, Cursor, or Claude Desktop, then restart the client.