[Enh]: `aggregate_records` should default fields to `*`

[JerryNixon]

If it is possible, it would align well.

tool

{
  "tools": [
    {
      "name": "aggregate_records",
      "inputSchema": {
        "type": "object",
        "properties": {
          "field": {
            "type": "string",
            "description": "Field name to aggregate, or '*' with count to count all rows."
          },
       }
     }
    }
  ]
}

desired state

{
  "tools": [
    {
      "name": "aggregate_records",
      "inputSchema": {
        "type": "object",
        "properties": {
          "field": {
            "type": "string",
            "description": "Field name to aggregate, or '*' with count to count all rows.",
            "default": "*" // new default value
          },
       }
     }
    }
  ]
}

[souvikghosh04]

This seems like a feature request, not a bug, since the original spec- #3178 didn't have "default" "*".

The current behavior of aggregate_records is working as designed:

• The field parameter is intentionally listed in the required array ("required": ["entity", "function", "field"]), meaning callers must explicitly provide it.
• When "*" is passed with count, it correctly counts all rows. When a specific field name is passed, it correctly aggregates on that field. No incorrect behavior exists.

What the issue requests:
Adding "default": "" to the field property so that omitting field defaults to ""

Implementation considerations:

• Add "default": "*" to the field property in the JSON schema.
• Remove "field" from the required array.
• Add fallback logic in the execution path to default to "*" when field is not provided.
• Validate that defaulting to "" only applies to count — since avg, sum, min, max with "" are invalid, omitting field for those functions should still return a validation error.

[souvikghosh04]

@Aniruddh25 / @JerryNixon : as discussed, updated this as Feature