wcagcheckr

For AI-native developers

Hand your accessibility audit to the AI you already code with.

You’re already pairing with Claude or Cursor. wcagcheckr ships an MCP server (@wcag-checkr/mcp) that exposes the audit as JSON-RPC tools your LLM IDE can call directly — so it can audit a page, read the findings, and propose fixes inside the same loop. Plus AI per-finding judges that catch what rules engines can’t (semantic alt text, heading clarity, color-only meaning) and ready-to-paste fix prompts.

Install free See pricing

Most accessibility tools are a separate dashboard you context-switch to, copy findings out of, and translate by hand. If your workflow is AI-in-the-editor, the audit should be a tool your agent can call — not another tab.

The AI features run on your own Anthropic key, so you pay only for the calls you make — no metered passthrough.

Start free — the core audit and exports

Install free; the matrix, visualizers, and AI-prompt exports get you going.

  • Multi-state matrix (preset) + the full visualizer pack
  • AX-tree viewer — see exactly what assistive tech sees
  • Standard exports including ready-to-paste AI fix prompts
  • VS Code extension + DevTools panel
Install free

Deterministic AI, on your key

Every AI judgment runs at temperature 0 — the same screenshot and prompt produce the same verdict twice, so your audit doesn’t flip-flop between runs. And it’s your Anthropic key, so cost and data stay yours.

Give your AI editor an accessibility audit it can actually call. Free to start.

Install free

No checkout here — on purpose. Install free, run it on your real site, and unlock the rest inside the app when the evidence is worth it to you.

Not your situation?   I got a demand letter · I need defensible evidence · I run audits for clients