fetch_result_rows
Page through a parked query result by its result_id (returned by execute_sql when a result exceeds the inline preview). Reads the stored snapshot — the exact rows the preview described, at zero warehouse cost and with no re-execution drift. Use this ONLY when you need to see specific raw rows beyond the preview; if what you actually want is a different aggregation or filter, write new SQL instead — it is cheaper and better. A page is capped at 20 KB as well as 1000 rows, so it can return fewer than limit: always continue from next_offset, never offset + limit. Handles expire after 24 hours; an expired id means re-running the SQL for a fresh one.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"result_id": "<result_id>"
},
"name": "fetch_result_rows"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Rows to return, max 1000. Defaults to 100. A page is also capped by size, so it can return fewer: always continue from `next_offset`. | |
| offset | No | First row to return (0-based). Defaults to 0. | |
| result_id | Yes | The `result_id` an `execute_sql` response returned. | |
| question_id | No |