How to create nested JSON schema file from details stored in database table in SQL Server

I have table that stores the information about the columns that JSON schema file needs to hold like tag name, its SQL Server type etc.. see below.

I need to write SQL that will create JSON output. I was able to achieve this only when there is no nesting is available. But I am stuck because my json needs to be nested.

in below example, I can store nested node names like "Details:" and "Name:" in new column but not sure how to use them to look like this.

I did level 0 nesting with below query.

Select 
STRING_AGG(CAST(CONCAT(REPLICATE(' ', 0), '"', friendlyname, '":{"type":["', commontype, '"', CASE WHEN nullable = 1 THEN ',"null"' ELSE '' END, '], "sqltype":"', sqltype, 
CASE WHEN maxlength IS NULL AND precision IS NULL AND scale IS NULL THEN '' ELSE CONCAT('(', maxlength, CAST(precision AS VARCHAR(2)) + ',', scale, ')') END, '", "nullable":"', 
CASE WHEN nullable = 1 THEN 'true' ELSE 'false' END, '", "Description":"', description, '"}') AS VARCHAR(MAX)), 
CONCAT(',', CHAR(13), CHAR(10))) WITHIN GROUP (ORDER BY friendlyname)
FROM Mytable

Table Structure

Table Data

Required Output:

{*-- nest level 0*
  "AccountNo": {
    "type": [
      "number"
    ],
    "sqltype": "int",
    "nullable": "false",
    "Description": ""
  },
  "AccountRegisterDate": {
    "type": [
      "string"
    ],
    "sqltype": "date",
    "nullable": "false",
    "Description": ""
  },
  "AccountSum": {
    "type": [
      "number"
    ],
    "sqltype": "decimal(28,4)",
    "nullable": "false",
    "Description": ""
  },
  "Details": { *-- nest level 1*
    "Name": { *-- nest level 2*
      "FirstName": {
        "type": [
          "string"
        ],
        "sqltype": "varchar(100)",
        "nullable": "false",
        "Description": ""
      },
      "LastName": {
        "type": [
          "string"
        ],
        "sqltype": "varchar(100)",
        "nullable": "false",
        "Description": ""
      },
      "MiddleName": {
        "type": [
          "string"
        ],
        "sqltype": "varchar(100)",
        "nullable": "false",
        "Description": ""
      }
    }
  }
}