Skip to content

Model Context Protocol

WebGuard MCP server

Query your WebGuard monitors and analytics directly from an MCP-compatible client. The Streamable HTTP endpoint exposes read-only tools and follows your account’s visibility rules and permissions.

Endpoint

https://app.webguard.dev/api/mcp

Authentication

Create a WebGuard API key with the “Read analytics” ability and send it as a bearer token. Use the key only in a trusted MCP client and treat it like a password.

Connect a client

Add the URL below as an MCP server. Your client must support Streamable HTTP and send the Authorization header with every request.

{
  "url": "https://app.webguard.dev/api/mcp",
  "headers": {
    "Authorization": "Bearer YOUR_API_KEY"
  }
}

Available tools

list_monitorings
Lists monitors visible to your account with their ID, name, type, and status. The optional limit parameter ranges from 1 to 100 (default: 50).
get_monitoring_analytics
Returns analytics for a monitor visible to your account. Required: monitoring_id. Optional: days from 1 to 365 (default: 30). The response uses the same analytics as the WebGuard dashboard.

Example prompts

  • Which of my monitors are visible, and what is their current status?
  • Summarize the analytics for monitoring 42 from the last 30 days.

Access and privacy

The server uses the same monitoring visibility rules as WebGuard. Private monitors from other accounts remain hidden. Tools do not return monitoring targets or credentials and cannot change data. Use a key with minimal permissions and revoke it when it is no longer needed.

REST API reference

The WebGuard REST API is also available for direct HTTP integrations. Open documentation.