How can I count all the occurrences within my date_histogram?

I'm trying to get a total count of my aggregations based on group_by_asset's field:summary.severity. Below is my es query:

{
  "runtime_mappings": {
    // Your runtime mappings here
  },
  "aggs": {
    "histogram_data": {
      "date_histogram": {
        "field": "timestamp",
        "calendar_interval": "day",
        "extended_bounds": {
          "min": "2023-06-12T07:00:00.000Z",
          "max": "2024-06-11T19:11:32.532Z"
        }
      },
      "aggs": {
        "group_by_asset": {
          "multi_terms": {
            "terms": [
              { "field": "asset.id" },
              { "field": "summary.severity" }, // This is what I want
              { "field": "asset.name", "missing": "N/A" }
            ]
          }
        }
      }
    }
  },
  "query": {
    ...
  }
}

The result I'm getting is a breakdown for each day:

{
  "group_by_asset": {
    "buckets": [
      {
        "key_as_string": "...",
        "key": ["123", 2, "BLT"], // This is what I want
        "doc_count": 1
      }
    ],
    "sum_other_doc_count": 0,
    "doc_count_error_upper_bound": 0
  },
  "key_as_string": "2024-04-18T00:00:00.000Z",
  "key": 1713398400000,
  "doc_count": 1
},
"group_by_asset": {
  ...

What I'm interested in is this part: "key": ["123", 2, "BLT"], where I want to keep track of the counts for severity values 1 or 2 within the bucket. So this total will have:

how_many_ones : <number>,  
how_many_twos: <number>

I've tried reorganizing the aggregations, using multiple aggregations, moving them into each other. Any suggestions on how to structure the query to achieve this?

I might be still missing something in your question, but the most obvious solution would be to just add a terms aggregation for the summary.severity field on the same level as histogram:

{
  "runtime_mappings": {
    // Your runtime mappings here
  },
  "aggs": {
    "by_severity": {
      "terms": { "field": "summary.severity" }
    },
    "histogram_data": {
      "date_histogram": {
        "field": "timestamp",
        "calendar_interval": "day",
        "extended_bounds": {
          "min": "2023-06-12T07:00:00.000Z",
          "max": "2024-06-11T19:11:32.532Z"
        }
      },
      "aggs": {
        "group_by_asset": {
          "multi_terms": {
            "terms": [
              { "field": "asset.id" },
              { "field": "summary.severity" }, // This is what I want
              { "field": "asset.name", "missing": "N/A" }
            ]
          }
        }
      }
    }
  },
  "query": {
    ...
  }
}