Include MCP Server Name as part of the payload for `beforeMCPExecution` Cursor Hook

Feature Requests
1

Feature request for product/service

Cursor IDE

Describe the request

I have started setting up cursor hooks in my repository to set up usage metrics for the different MCP servers we have set up. To do this, I need a way of identifying which MCP server call triggered the hook. For servers defined with a url this is not too bad since each server has its url. However, when configuring with command, several of the servers I have set up use docker as the command which on its own is not enough to identify the server. Looking deeper, I was surprised to find that for beforeMCPExecution the inputs look like this but notably do not include the actual MCP server name.

{
  "conversation_id": "string",
  "generation_id": "string",
  "hook_event_name": "string",
  "workspace_roots": ["<path>"],
  "tool_name": "<tool name>",
  "tool_input": "<json params>"
}
// Plus either:
{ "url": "<server url>" }
// Or:
{ "command": "<command string>" }

Considering the hooks.json file already is set up as below, it makes the most sense that <mcp-server-name-1> and <mcp-server-name-2> also be passed down as part of the input to the hook.

{
  "mcpServers": {
    <mcp-server-name-1>: {
      "command": <command>,
      "args": [...],
      "env": {
        <env-var>: <value>
      }
    },
    <mcp-server-name-2>: {
      "url": <url>
    }
  }
}

Alternatively, passing in args instead of just command could be enough for me to identify which server it is but it makes the most sense to just include the mcp_server_name in the payload.