Integrations

MCP Server

Use SchemaCheck natively inside any Model Context Protocol compatible AI tool. Validate URLs and raw JSON-LD without leaving your editor.

Claude DesktopCursorWindsurfVS Code + ClineVS Code + ContinueZedFirebase StudioGemini CLI

Two tools, one integration

validate_schema

Validate any URL or raw JSON-LD object. Returns errors, warnings, rich result eligibility, fix suggestions, and links to Google's documentation.

list_supported_types

Returns all 35+ schema types with their validation status and Google docs links.

Prerequisites

✓Node.js 18+ installed
✓A SchemaCheck API key — get one free
✓Any MCP-compatible AI client (see list above)

Claude Desktop

Open ~/Library/Application Support/Claude/claude_desktop_config.json and add:

json

{
  "mcpServers": {
    "schemacheck": {
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Replace YOUR_API_KEY with your key from the dashboard. Restart Claude Desktop to activate.

Cursor

Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):

json

{
  "mcp": {
    "servers": {
      "schemacheck": {
        "command": "npx",
        "args": ["-y", "schemacheck-mcp"],
        "env": {
          "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
        }
      }
    }
  }
}

Windsurf

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

json

{
  "mcpServers": {
    "schemacheck": {
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Windsurf's Cascade agent picks up MCP servers automatically after saving. No restart required.

VS Code + Cline

Open the Cline panel → click MCP Servers → Configure MCP Servers. This opens cline_mcp_settings.json. Add:

json

{
  "mcpServers": {
    "schemacheck": {
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
      },
      "disabled": false,
      "alwaysAllow": ["validate_schema", "list_supported_types"]
    }
  }
}

alwaysAllow auto-approves these tools so Cline won't prompt on every call.

VS Code + Continue

Create .continue/mcpServers/schemacheck.yaml in your project root:

yaml

name: schemacheck
command: npx
args:
  - "-y"
  - schemacheck-mcp
env:
  SCHEMACHECK_API_KEY: YOUR_API_KEY

Continue auto-detects any YAML or JSON file placed in the .continue/mcpServers/ directory.

Zed

Open ~/.config/zed/settings.json and add a context_servers key:

json

{
  "context_servers": {
    "schemacheck": {
      "source": "custom",
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Zed restarts the context server automatically whenever you save settings.json.

Firebase Studio

Create .idx/mcp.json at your project root. Use ${env:VAR} syntax to reference environment variables from your .env file:

json

{
  "mcpServers": {
    "schemacheck": {
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "${env:SCHEMACHECK_API_KEY}"
      }
    }
  }
}

Add SCHEMACHECK_API_KEY=your_key to your .env file at the project root. MCP support in Firebase Studio is currently in preview.

Gemini CLI

Add to ~/.gemini/settings.json (global) or .gemini/settings.json (project):

json

{
  "mcpServers": {
    "schemacheck": {
      "command": "npx",
      "args": ["-y", "schemacheck-mcp"],
      "env": {
        "SCHEMACHECK_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Manage servers from the CLI with gemini mcp enable schemacheck / gemini mcp status.

Usage examples

Validate a live URL

javascript

// Ask your AI assistant:
"Validate the schema markup on https://example.com"

// The MCP tool call looks like:
validate_schema({ url: "https://example.com" })

Validate raw JSON-LD

javascript

// Ask your AI assistant:
"Check if this schema is valid for rich results"

// The MCP tool call looks like:
validate_schema({
  jsonld: {
    "@context": "https://schema.org",
    "@type": "Article",
    "headline": "My Article",
    "author": { "@type": "Person", "name": "Jane Doe" },
    "datePublished": "2024-01-01"
  }
})

List supported types

javascript

// Ask your AI assistant:
"What schema types does SchemaCheck support?"

// The MCP tool call looks like:
list_supported_types()

What your AI assistant can do

Audit any page

"Check the schema markup on our homepage and list all errors"

Fix schema inline

"Validate this JSON-LD I wrote and fix any missing required fields"

Pre-publish checks

"Before I deploy this product page, validate its schema against Google's requirements"

Bulk audits

"Validate all 10 URLs from this list and summarize the errors"

Rich result debugging

"Why isn't this Article schema eligible for rich results?"

Schema generation

"Generate valid Article schema for this blog post and confirm it passes"

Point to a custom API base

By default the server calls https://schemacheck.dev/api/v1. Override it for self-hosted or local instances:

json

"env": {
  "SCHEMACHECK_API_KEY": "YOUR_API_KEY",
  "SCHEMACHECK_API_BASE": "http://localhost:3000/api/v1"
}

Plan limits apply

Each validate_schema call counts as one validation against your monthly quota. Cached results (same URL within 1 hour) are free and don't count. View plans →