datarails-finance-os
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Endpoint
Hosted endpoint · transport http
Connect
This is a remote MCP server — point an MCP client at the endpoint URL. Clients with native remote support (Claude, Cursor, VS Code, …):
For stdio-only clients (older Claude Desktop, etc.), bridge with mcp-remote:
If the server requires authentication, add the provider's token/header (e.g. an X-API-Key) per its docs. A quick reachability check:
Tools
list_data_models— Entry point for both raw data layers. Lists available finance tables / data models; each entry carries the numeric id (for the by-id tools) and the alias (for the by-alias tools; empty when a table has no alias).list_aliased_fields— Lists the aliased (friendly-named) fields of a table. Preferred raw-data entry point — roughly 95% fewer tokens than the by-id layer. A table alias does not imply its fields are aliased.get_fields_by_id— Returns column names, types and relationships for a table by numeric id.list_business_metrics— Lists the org's named business metrics (KPI catalog). Ungated, so it is the discovery entry point for the business-metrics layer even where the metric-data tools are feature-flag denied.profile_numeric_fields— Baseline numeric statistics (SUM/AVG/MIN/MAX/COUNT) per field, relayed in the backend-native DR_Values / col_keys / row_keys layout.profile_categorical_fields— Category frequencies and cardinality. Must be given an explicit business-dimension field list, or it profiles upload/mapping metadata columns rather than business data.get_data_by_alias— Retrieve filtered rows from an aliased table (max 500 rows per page).get_data_by_id— Retrieve filtered rows from a table by numeric field ids (max 500 rows per page).start_aggregation_by_alias— Start a grouped aggregation over an aliased table. Non-blocking; returns {status: pending, handle}. No row limit on the result.get_aggregation_result_by_alias— Poll for the result of start_aggregation_by_alias. status "running" plus retry_after_seconds means poll again; it is not an error.start_aggregation_by_id— Start a grouped aggregation over a table addressed by numeric field ids.get_aggregation_result_by_id— Poll for the result of start_aggregation_by_id.get_aggregated_data_by_alias— Blocking aggregation over an aliased table.get_aggregated_data_by_id— Blocking aggregation over a table by numeric field ids.start_distinct_values_by_alias— Start a distinct-values fetch for an aliased field. Non-blocking.get_distinct_values_result_by_alias— Poll for distinct values started by start_distinct_values_by_alias; takes the limit argument.start_distinct_values_by_id— Start a distinct-values fetch for a field addressed by numeric id.get_distinct_values_result_by_id— Poll for distinct values started by start_distinct_values_by_id.get_distinct_values_by_alias— Blocking distinct-values fetch for an aliased field.get_distinct_values_by_id— Blocking distinct-values fetch for a field by numeric id.get_business_metric_details— Definition and metadata for one named business metric.get_business_metric_data— Values for one named business metric.get_business_metric_drilled_down_data— Drill a business metric down by a dimension.get_business_metric_table_rows— Underlying rows behind a business metric.sql_query— Free-form SQL over the org's finance tables.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.