Requete Tableur (Langage Naturel)
Queries a spreadsheet using natural language. The system translates the question into a structured query and returns matching results.
Parameters
| Parameter | Type | Required | Variable | Description |
|---|---|---|---|---|
query | dynamic value | Yes | Yes | Natural language question to query the spreadsheet. The system interprets the question and generates the query automatically. |
spreadsheetIds | array | No | No | Spreadsheet identifiers to query. Empty to search across all workspace spreadsheets. |
maxResults | number | No | No | Maximum number of rows in the results. (Default: 100) |
includeExplanation | boolean | No | No | Include the explanation of the query interpreted by the model. (Default: false) |
llmSynthesis | object | No | No | LLM result synthesis configuration (enable and custom prompt). |
outputVariable | text | No | No | Output variable name containing the query results. |
Parameters marked Variable = Yes accept the
{{blockName.field}}syntax.
Output
Output variable : nlQueryResult
{
"results": [],
"total": 0,
"has_more": false,
"success": false,
"explanation": {},
"aggregations": {}
}
Example
Ask a question about a spreadsheet.
Input :
{"query": "Quels sont les clients avec un CA superieur a 100k ?"}
Output :
{"results": [{"name": "Acme", "revenue": 250000}], "count": 1}
Tip
Formulate clear and specific questions. The block uses an LLM to interpret the question and generate the query.