Retrieve accessibility issue data directly into your integrated development environment (IDE) or AI assistant. This integration supports faster and more efficient remediation within your existing development workflows.
Note: For additional details about Model Context Protocol (MCP), visit our Level Access Developer Documentation Help Center.
On this page:
- Set up an API key in the platform
- Configuration prerequisites
- Configure Visual Studio Code for MCP
- Configure Windsurf for MCP
- Configure JetBrains WebStorm for MCP
- Configure Claude Code for MCP
- Configure Cursor and other MCP clients
- Example use cases and prompts
- Related resources
Set up an API key in the platform
Log in to the Level Access Platform and create an API key. You need this key to authenticate requests between your IDE or AI assistant and the MCP server.
- Log in to the platform.
- From the Control hub, go to Tools & Integrations, then select API.
- Select +Add new API key and generate your token.
- Copy and securely store your API token in a password manager.
For detailed step-by-step instructions, see Create an API token to use Level Access API.
Configuration prerequisites
When configuring any MCP client or IDE, you must provide the following connection details:
-
API key header (
x-level-access-api-key): The API token generated from Tools & Integrations. -
Tenant domain header (
x-level-access-tenant-domain): Your organization's specific subdomain name (for example,myorgif your platform address ismyorg.levelaccess.net). Do not include the full URL or protocol. - Transport type: HTTP.
-
Endpoint URL:
https://mcp.hub.essentia11y.com/mcp(orhttps://mcp.eu.levelaccess.net/mcpfor organizations hosted in the European Union environment).
Configure Visual Studio Code for MCP
Add the Level Access MCP configuration to your Visual Studio Code settings to connect your editor to the platform:
- Open Visual Studio Code.
- Open the Command Palette by pressing Ctrl+Shift+P (Windows/Linux) or Cmd+Shift+P (macOS).
- Type and select Preferences: Open Settings (JSON).
-
Add the following configuration to your settings file:
{ "servers": { "Level Access MCP": { "type": "http", "url": "https://mcp.hub.essentia11y.com/mcp", "headers": { "x-level-access-api-key": "your-api-key", "x-level-access-tenant-domain": "your-tenant-domain" } } } } - Replace
your-api-keywith your actual API key. - Replace
your-tenant-domainwith your organization subdomain. - Save the file and restart Visual Studio Code.
Note: In the
x-level-access-tenant-domainline, don't include the full URL. Enter only your organization's specific subdomain name.
Configure Windsurf for MCP
If you use Windsurf, configure the Level Access MCP service using the Cascade MCP configuration file:
- Open Windsurf.
- Open your Cascade settings, or open your global MCP configuration file (
~/.codeium/windsurf/mcp_config.json). -
Add the following configuration under the
mcpServersobject:{ "mcpServers": { "level-access": { "serverUrl": "https://mcp.hub.essentia11y.com/mcp", "headers": { "x-level-access-api-key": "your-api-key", "x-level-access-tenant-domain": "your-tenant-domain" } } } } - Replace
your-api-keywith your actual API key. - Replace
your-tenant-domainwith your organization subdomain. - Save your changes and restart Windsurf.
Note: In the
x-level-access-tenant-domainline, don't include the full URL. Enter only your organization's specific subdomain name.
Configure JetBrains WebStorm for MCP
If you use JetBrains WebStorm or another JetBrains IDE with an MCP integration plugin, use the standard mcpServers object format:
- Open WebStorm.
- Open your IDE settings and locate your MCP configuration settings.
-
Add the following configuration:
{ "mcpServers": { "Level Access MCP": { "type": "http", "url": "https://mcp.hub.essentia11y.com/mcp", "headers": { "x-level-access-api-key": "your-api-key", "x-level-access-tenant-domain": "your-tenant-domain" } } } } - Replace
your-api-keywith your actual API key. - Replace
your-tenant-domainwith your organization subdomain. - Save your changes and restart your IDE.
Configure Claude Code for MCP
To add the Level Access MCP server to Claude Code from the command line:
- Open your terminal application.
-
Run the following command:
claude mcp add --transport http lap https://mcp.hub.essentia11y.com/mcp --header "x-level-access-api-key: your-api-key" --header "x-level-access-tenant-domain: your-tenant-domain"
- Replace
your-api-keywith your actual API key andyour-tenant-domainwith your organization subdomain. - Verify the connection by running
claude mcp list.
Configure Cursor and other MCP clients
For Cursor and other AI-assisted developer tools that support MCP configuration through JSON files, you can use either an array or an object structure depending on your client requirements.
To configure clients that use an array of servers:
{
"mcpServers": [
{
"name": "level-access",
"transport": "http",
"url": "https://mcp.hub.essentia11y.com/mcp",
"headers": {
"x-level-access-api-key": "your-api-key",
"x-level-access-tenant-domain": "your-tenant-domain"
}
}
]
}To configure clients that use an object map of server names:
{
"mcpServers": {
"level-access": {
"type": "http",
"url": "https://mcp.hub.essentia11y.com/mcp",
"headers": {
"x-level-access-api-key": "your-api-key",
"x-level-access-tenant-domain": "your-tenant-domain"
}
}
}
}Note: Replace
your-api-keywith your actual API key andyour-tenant-domainwith your organization subdomain. Your specific IDE or client may require a slightly different schema.
Example use cases and prompts
The Level Access MCP server enables AI assistants to retrieve accessibility data from the Level Access Platform, making it easier to analyze, report on, and remediate accessibility issues.
Accessibility issue analysis during development
While working on a website or application in your IDE, ask your AI assistant to analyze accessibility issues directly in your codebase:
Example prompt:
"What are the top accessibility issues on our main website? Show me the critical severity issues and their details."
What the AI assistant does:
- Retrieves your digital assets using
get_digital_assets. - Fetches issues for your main website with
get_digital_asset_issues. - Filters and displays critical issues with full details via
get_digital_asset_issue_details. - Provides remediation guidance based on rule details from
get_rule.
Progress tracking and reporting
When preparing status reports, sprint reviews, or audit preparations:
Example prompt:
"Generate a summary of our accessibility metrics across all workspaces and compare the issue counts by severity."
What the AI assistant does:
- Lists all workspaces with
get_workspaces. - Retrieves metrics for each workspace using
get_platform_metrics. - Creates a formatted summary showing severity distribution and trends across critical, high, low, and info levels.
- Highlights digital assets and areas that need immediate attention.
Targeted remediation and code fixes
When addressing a specific finding on a component or page:
Example prompt:
"How do I fix the color contrast and missing alternative text issues identified in our digital asset? Provide accessible code examples."
What the AI assistant does:
- Queries the finding details and affected code elements.
- Retrieves the associated accessibility rule recommendations and WCAG success criteria.
- Suggests compliant HTML, CSS, or ARIA implementations directly in your editor.
Comments
0 comments
Article is closed for comments.