Skip to content
MCP server

AccessibilityBrief

WCAG 2.2 scan results and fix prompts for AI coding agents, with rescan to confirm

Runs locallyStatus and history
Install
Pricing
Freemium
Platforms
CLI, API
Runs
Locally
Version
1.1.2
Listing updated
2 days ago

About AccessibilityBrief

AccessibilityBrief is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.

The loop here is scan, fix, rescan. An agent reads the latest accessibility scan for a site, takes the worst issue group, fetches the fix prompt with selectors, code and expected result, edits your repository, and after you deploy it starts a new scan to check. Scans test your live or staging site rather than local files, so a fix only shows once deployed. The agent cannot mark an issue as fixed itself; the next scan detects that.

The server runs locally through npx -y accessibilitybrief mcp and needs Node.js 18.17 or newer. It requires an API key from Settings with read scope, plus scan to start scans and write to mark issues as false positive, won't fix or needing a manual check. The site must already be in your workspace, and API access is part of the Starter, Pro and Agency plans.

The docs caution that automated testing finds only part of WCAG issues, so keyboard and screen reader checks are still needed.

Install

Run in a terminal

claude mcp add mcp --env ACCESSIBILITYBRIEF_API_KEY='<ACCESSIBILITYBRIEF_API_KEY>' -- npx -y accessibilitybrief

Add to claude_desktop_config.json

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "accessibilitybrief"
      ],
      "env": {
        "ACCESSIBILITYBRIEF_API_KEY": "<ACCESSIBILITYBRIEF_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "accessibilitybrief"
      ],
      "env": {
        "ACCESSIBILITYBRIEF_API_KEY": "<ACCESSIBILITYBRIEF_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "accessibilitybrief"
      ],
      "env": {
        "ACCESSIBILITYBRIEF_API_KEY": "<ACCESSIBILITYBRIEF_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "accessibilitybrief"
      ],
      "env": {
        "ACCESSIBILITYBRIEF_API_KEY": "<ACCESSIBILITYBRIEF_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mcp": {
      "command": "npx",
      "args": [
        "-y",
        "accessibilitybrief"
      ],
      "env": {
        "ACCESSIBILITYBRIEF_API_KEY": "<ACCESSIBILITYBRIEF_API_KEY>"
      }
    }
  }
}

Configuration

  • ACCESSIBILITYBRIEF_API_KEYAPI key from AccessibilityBrief Settings → API and integrations (Starter, Pro or Agency plan). Scopes: read; scan to start scans; write to mark issues.Required

Tools this server provides (7)

Taken from the project documentation; the running server may differ.

list_sites
Lists sites in the workspace with their latest score.
get_latest_scan
Reads the latest finished scan of a site.
get_issue_groups
Lists issue groups worst first with affected elements, WCAG criteria and fix steps.
get_fix_prompt
Fetches the fix prompt for one issue group or a whole site.
start_scan
Starts a new scan.
get_scan
Checks scan progress and reads its result.
mark_issue_status
Marks issues as false positive, won't fix or needing a manual check.

What it does

Scan reading
Reads the latest finished scan with score, severity counts and change since the previous scan.
Issue groups
Lists groups worst first with affected elements, WCAG criteria and fix steps.
Fix prompts
Fetches a prompt for one group or a whole site with selectors and expected result.
Rescan on demand
Starts a new scan after a fix is deployed, using one manual scan of the plan.
Issue triage
Marks issues as false positive, won't fix or needs manual check, with a note.
Scoped API key
Separate read, scan and write scopes limit what the agent can do.

Good for

  • Fix the worst accessibility issue group in a repo with an agent
  • Confirm that a deployed fix removed an issue
  • Triage false positives with a note
  • Check a site's score after a release

Pricing

Freemium, from $19 a month

Free single-page check; MCP access on paid plans from $19/month

Prices change. Check accessibilitybrief.com before you buy.

AccessibilityBrief FAQ

Which plans include the MCP server?

Starter, Pro and Agency.

What do I need to run it?

Node.js 18.17 or newer and an API key with the scopes you want to grant.

Can the agent mark issues fixed?

No. Fixed issues are detected by the next scan.

Does it scan local files?

No. Scans check your live or staging site, so fixes appear after deployment.

Does it find every WCAG issue?

No. Automated testing finds only part of them; manual checks are still needed.

Is AccessibilityBrief free?

AccessibilityBrief has a free plan, with paid plans for more usage or features. Paid plans start at $19 a month.

How do I add AccessibilityBrief to Claude Code?

Run this command in your terminal: claude mcp add mcp --env ACCESSIBILITYBRIEF_API_KEY='<ACCESSIBILITYBRIEF_API_KEY>' -- npx -y accessibilitybrief

What are the best alternatives to AccessibilityBrief?

The closest alternatives to AccessibilityBrief are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to AccessibilityBrief

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools