get_dimensions
Extract dimensions for specified metrics to categorize and analyze data attributes effectively. Ideal for identifying key characteristics in datasets.
Instructions
Dimensions are the attributes, features, or characteristics that describe or categorize data.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| metrics | Yes |
Implementation Reference
- Primary MCP tool handler for 'get_dimensions'. This is the function invoked when the tool is called, which delegates to the SemanticLayerFetcher.
@dbt_mcp_tool( description=get_prompt("semantic_layer/get_dimensions"), title="Get Dimensions", read_only_hint=True, destructive_hint=False, idempotent_hint=True, ) async def get_dimensions( context: SemanticLayerToolContext, metrics: list[str], search: str | None = None ) -> list[DimensionToolResponse]: return await context.semantic_layer_fetcher.get_dimensions( metrics=metrics, search=search ) - Core helper function in SemanticLayerFetcher that performs the actual GraphQL query to fetch dimensions and caches the results.
async def get_dimensions( self, metrics: list[str], search: str | None = None ) -> list[DimensionToolResponse]: metrics_key = ",".join(sorted(metrics)) if metrics_key not in self.dimensions_cache: dimensions_result = await submit_request( await self.config_provider.get_config(), { "query": GRAPHQL_QUERIES["dimensions"], "variables": { "metrics": [{"name": m} for m in metrics], "search": search, }, }, ) dimensions = [] for d in dimensions_result["data"]["dimensionsPaginated"]["items"]: dimensions.append( DimensionToolResponse( name=d.get("name"), type=d.get("type"), description=d.get("description"), label=d.get("label"), granularities=d.get("queryableGranularities") + d.get("queryableTimeGranularities"), ) ) self.dimensions_cache[metrics_key] = dimensions return self.dimensions_cache[metrics_key] - Pydantic/dataclass defining the output schema for dimension responses.
class DimensionToolResponse: name: str type: DimensionType description: str | None = None label: str | None = None granularities: list[str] | None = None - src/dbt_mcp/semantic_layer/tools.py:157-164 (registration)List of semantic layer tools including get_dimensions, used for registration into the MCP server.
SEMANTIC_LAYER_TOOLS = [ list_metrics, list_saved_queries, get_dimensions, get_entities, query_metrics, get_metrics_compiled_sql, ] - src/dbt_mcp/tools/toolsets.py:43-46 (registration)Toolset definition including get_dimensions for enable/disable configuration.
Toolset.SEMANTIC_LAYER: { ToolName.LIST_METRICS, ToolName.LIST_SAVED_QUERIES, ToolName.GET_DIMENSIONS,