datatomcp
datatomcp hosts your tabular data for Claude. Upload a CSV, paste a CSV link, or pick a Google Sheet (private sheets work through Google's file picker, which grants access to that file only). Every import comes with a report: unparseable lines with their line numbers, numbers stored as text, mixed date formats. Nothing is converted until you confirm. Claude gets five read-only tools: list_datasets, describe_dataset, query_rows, aggregate and distinct_values. Totals and counts are computed by a database (DuckDB), so Claude reports numbers instead of estimating them from pasted text, and a 50,000-row sheet that is too big to paste into a chat can still be queried. Every result names the dataset version it came from. Free: 3 datasets, 100,000 rows each. Pro ($12 or €12 a month): 25 datasets, 2M rows each, hourly Google Sheets sync. Data is stored in the EU, read-only, and never sent to AI providers or used for training.
Choose a connection option
5 tool names reported
Source-declared tools. Availability can depend on the connection and permissions; these are not runtime test results.
Claude connector directory ↗ · Observed 2026-11-01 · Completeness not established
aggregatedescribe_datasetdistinct_valueslist_datasetsquery_rows
Sources and provenance
- Claude connector directoryClaude tier: community · Observed
- Official MCP RegistryObserved
Grouping basis: endpoint. Exact endpoints or package sets can be grouped automatically; other relationships require evidence review. This does not verify publisher affiliation, connection health, or equivalent capabilities.