mcpdata Request an invite

Answers / Get data into your AI

How do I let Claude query a CSV file on my computer?

Recipe verified on Ran on a private test instance
Short answer

Sync the folder with the CSV to an mcpdata instance with cdb-localdata, and connect Claude to that instance. Claude then queries the file as a table. The file is only read, and it stays in step: run the sync again (or install it to run every minute) and changes arrive as a new version.

Steps

  1. Connect your assistant to your instance (this writes, so it needs your own instance with a ReadWrite key from the accounts portal; the public demo is read-only). In Claude Code: 
    claude mcp add --transport http mcpdata https://[you].mcpdata.org/mcp --header "Authorization: Bearer $CDB_API_KEY"

    Replace [you] with your instance's name. Other apps: Connect your AI.

  2. Install the client and sign it in.
    pip install https://[you].mcpdata.org/help/download/cdb-0.3.4-py3-none-any.whl
cdb-localdata login --to https://[you].mcpdata.org
  3. Add the folder and sync it. Here a folder mixed with the sample orders_mixed_dates.csv: 
    cdb-localdata add ./samples/mixed --to https://[you].mcpdata.org --namespace mycsv
cdb-localdata sync
  4. Keep it in step (optional). Sync every minute in the background, started with your computer (Task Scheduler, launchd or cron): 
    cdb-localdata install
  5. Ask your question. Give your AI this prompt:
    Using mcpdata, how many orders per channel are in mycsv.orders_mixed_dates?
What your AI does: the tool call it makes for this prompt, and what came back
query(table_name="mycsv.orders_mixed_dates", group_by=["channel"], agg=[{"func": "count", "column": "order_id", "alias": "orders"}], sort=[{"column": "orders", "desc": true}]) → 3 rows
What you should get · the recipe's real result from its last run
channelorders
Online5,507
Store3,224
Partner1,269
Ask Claude with this prompt Ask ChatGPT with this prompt

The assistant needs mcpdata connected (step 1) to run the query.

Why this works

Claude can't open files on your disk by itself, and pasting a CSV into the chat runs into the context limit. Syncing the folder makes the file a table Claude can query with exact results, and your other tools (Python, JDBC, the workbench) see the same table.

Related questions