A11Y Toolkit
WCAG 2.2 audit, fix and evidence tools for AI coding agents
- Pricing
- Open source
- Platforms
- CLI, Web
- Runs
- Locally
- Version
- 5.27.0
- Listing updated
- 3 days ago
About A11Y Toolkit
A11Y Toolkit is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.
The toolkit pairs an MCP server (37 tools, 5 prompts and a skill) with a CLI that shares the same engine. Static audits, contrast math, PDF and DOCX screening and the evidence pack run with zero dependencies; rendered checks such as real contrast over images, keyboard traps, reflow and forms need Playwright and a browser.
It claims automated signals for 51 of 55 A/AA criteria and attaches a concrete remediation to every finding. Its stance is that a machine finds issues and a person signs: the maker says automated scanning covers roughly a third of WCAG and reports show that boundary. A blindfold usability test lets an agent attempt tasks using only what a screen reader exposes.
The companion service, a11y.es, schedules weekly audits and issues hash-verifiable monthly evidence packs, aimed at Spanish agencies and shops covered by the European Accessibility Act.
Install
Run in a terminal
claude mcp add a11y-toolkit -- npx -y @kinti/a11y-toolkitAdd to claude_desktop_config.json
{
"mcpServers": {
"a11y-toolkit": {
"command": "npx",
"args": [
"-y",
"@kinti/a11y-toolkit"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"a11y-toolkit": {
"command": "npx",
"args": [
"-y",
"@kinti/a11y-toolkit"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"a11y-toolkit": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@kinti/a11y-toolkit"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"a11y-toolkit": {
"command": "npx",
"args": [
"-y",
"@kinti/a11y-toolkit"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"a11y-toolkit": {
"command": "npx",
"args": [
"-y",
"@kinti/a11y-toolkit"
]
}
}
}What it does
- WCAG 2.2 loop
- Audit, fix, document, watch and hand off in one toolset.
- Concrete remediations
- Every finding includes a fix the agent can apply.
- Evidence packs
- SHA-256 verifiable reports with an optional human signature.
- Blindfold testing
- An agent attempts tasks through screen-reader-visible output only.
- Bilingual engine
- Spanish and English, with MIT-licensed open-source code.
Good for
- Audit a page's accessibility from inside a coding agent
- Generate an evidence pack for a client or inspector
- Check colour contrast pairs quickly
- Catch regressions before a release
Pricing
Open source
MIT engine; a11y.es hosted service pricing not listed
Prices change. Check a11y.es before you buy.
A11Y Toolkit FAQ
Does it need dependencies?
Static audits and contrast math have none; rendered tools need Playwright.
Does it send telemetry?
The site states zero telemetry.
Can it prove a site is fully accessible?
No; it says machines see about a third of WCAG and a human must sign.
Is A11Y Toolkit free?
Yes. A11Y Toolkit is open source, so you can use and self-host it for free.
How do I add A11Y Toolkit to Claude Code?
Run this command in your terminal: claude mcp add a11y-toolkit -- npx -y @kinti/a11y-toolkit
What are the best alternatives to A11Y Toolkit?
The closest alternatives to A11Y Toolkit are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to A11Y Toolkit

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools