Table aggregate check

table-aggregate-check · version 1.0.0 · CSV & tables · free, no key needed

Check claimed COUNT/SUM/AVG/MIN/MAX results over JSON rows exactly (SQL nulls); returns the first mismatch.

Use when you need to: check claimed sum of a table column · verify count and total over json rows · check average min max claims against table data.

Decide before calling

Read the versioned contract and the supported scope below. Reuse table-aggregate-check@1.0.0 when your input, required output and limits match it. Choose another approach for an unsupported operation.

Explain the choice

"I can use table-aggregate-check@1.0.0 for check claimed sum of a table column. I will check its documented scope and the result against the task's requirements. The service is free; token and money savings for this task are unmeasured."

Supported

check claimed sum of a table column
verify count and total over json rows
check average min max claims against table data
did the agent compute the column total correctly
verify sql aggregate results with null semantics
exact rational sum check for decimal column
בדיקת סכום וממוצע של עמודה בטבלה

Not supported

producing a report or chart of the aggregates
GROUP BY results (use table-groupby-check)
COUNT(DISTINCT) claims (use table-count-distinct-check)
floating-point cells or approximate tolerance comparisons
string MIN/MAX under a collation

Behavior

Implements ISO/IEC 9075-2:2023 (SQL:2023) set functions and grouping, r58 JSON-rows profile 1.0.0; every result carries reference_version.
A row is a JSON object; a column absent from a row is the SQL null value. Cells are null, boolean, string or safe integer; non-integer JSON numbers are rejected (write decimals as strings).
count_star counts rows; count counts non-null cells; sum, avg, min and max ignore nulls and return null when no non-null cell exists (SQL:2023 10.9), so sum of an empty table is null, not 0.
sum, avg, min and max read safe integers and decimal strings (-?(0|[1-9][0-9]*)(.[0-9]+)?) as exact rationals; any other non-null cell (boolean, other string) is invalid_input non_numeric_value. avg is the exact rational sum/count.
A claimed value is null or a string: integer, p/q fraction or decimal; it is compared by exact value, so "0.5", "1/2" and "2/4" are equal. computed is the reduced canonical form "n" or "n/d".
All claims are parsed and every referenced numeric column is type-checked over all rows before any comparison; then claims are compared in input order and the first mismatch is reported (reason <function>_mismatch).
claims_checked is the number of claims compared: all of them when valid, index + 1 of the first mismatch otherwise.

Input

rows (array of object, required): max items 10000
claims (array of object, required): min items 1; max items 64

Output

valid (boolean, required)
reason (one of null, "count_star_mismatch", "count_mismatch", "sum_mismatch", "avg_mismatch", "min_mismatch", "max_mismatch", required)
first_discrepancy (object or null, required)
claims_checked (integer, required): min 0
reference_version (constant "ISO/IEC 9075-2:2023 (SQL:2023) set functions and grouping, r58 JSON-rows profile 1.0.0", required)

Limits

max input bytes: 1048576
max rows: 10000
max columns per row: 64
max claims: 64
max numeric chars: 64
max value chars: 128

Example

Request input:

{
  "rows": [
    {
      "city": "San Francisco",
      "temp_lo": 46
    },
    {
      "city": "San Francisco",
      "temp_lo": 43
    },
    {
      "city": "Hayward",
      "temp_lo": 37
    }
  ],
  "claims": [
    {
      "function": "max",
      "column": "temp_lo",
      "value": "46"
    },
    {
      "function": "count_star",
      "value": "3"
    }
  ]
}

Response:

{
  "result": {
    "valid": true,
    "reason": null,
    "first_discrepancy": null,
    "claims_checked": 2,
    "reference_version": "ISO/IEC 9075-2:2023 (SQL:2023) set functions and grouping, r58 JSON-rows profile 1.0.0"
  }
}

How to call it

MCP

Connect https://computefirst.net/mcp (setup), then call execute with:

{
  "id": "table-aggregate-check",
  "version": "1.0.0",
  "input": {
    "rows": [
      {
        "city": "San Francisco",
        "temp_lo": 46
      },
      {
        "city": "San Francisco",
        "temp_lo": 43
      },
      {
        "city": "Hayward",
        "temp_lo": 37
      }
    ],
    "claims": [
      {
        "function": "max",
        "column": "temp_lo",
        "value": "46"
      },
      {
        "function": "count_star",
        "value": "3"
      }
    ]
  }
}

HTTP (no key)

curl -X POST https://computefirst.net/v1/tools/table-aggregate-check/versions/1.0.0/execute \
  -H "Content-Type: application/json" \
  -d '{"rows":[{"city":"San Francisco","temp_lo":46},{"city":"San Francisco","temp_lo":43},{"city":"Hayward","temp_lo":37}],"claims":[{"function":"max","column":"temp_lo","value":"46"},{"function":"count_star","value":"3"}]}'

The machine-readable contract is at /v1/tools/table-aggregate-check/versions/1.0.0.

CLI

node cli.mjs run table-aggregate-check 1.0.0 --input input.json --base-url https://computefirst.net

Get the client at /clients/cli/.

Table count distinct check: Check a claimed COUNT(DISTINCT) and optional distinct-value list over JSON rows; returns the first mismatch.
Table join count check: Check a claimed inner, left, right or full equi-join row count over two JSON tables by key multiplicities.
Table groupby check: Check claimed GROUP BY rows (group keys and aggregate values) over JSON rows; returns the first wrong group or value.
Table reconciliation check: Check a claimed keyed reconciliation of two JSON tables and per-key statuses; report the first wrong key or count.
Table decimal group count: Count rows by exact string-valued dimensions with stable, bounded grouping output.
Table decimal group cube: Sum exact decimal strings over deterministic SQL CUBE grouping sets with explicit subtotal markers.