Dataloader Server

Overview

Dataloader retrieves market price history, adds technical indicators, and exports tables for other research servers. Fetching data and exporting a file are separate operations.

Connection

MCP URL: https://quantx-api.limex.pro/mcps/dataloader Select Dataloader in the QuantX marketplace. Use its gateway connection with Streamable HTTP and QuantX authorization; follow Client Setup. Yahoo requests need no provider key. For Massive, supply your own Massive-API-Key header; a configured service key may also be available in a particular deployment. Set source explicitly for reproducibility instead of relying on the provider default. Cursor configuration (.cursor/mcp.json):

{
  "mcpServers": {
    "dataloader": {
      "url": "https://quantx-api.limex.pro/mcps/dataloader",
      "headers": {
        "Authorization": "Bearer ${env:QUANTX_MCP_TOKEN}"
      }
    }
  }
}

Set the referenced environment variables before starting Cursor. For Claude Code or Python, use the client-specific format in Client Setup.

Tools

Tool	Inputs	Result
`fetch-ohlcvs`	Required `tickers`, `start_date`, `end_date`, and `timeframe` (or legacy `interval`); optional `source`	`ohlcv_df`, `ticker_details`, `row_count`, `columns`, or a stored reference
`list-supported-indicators`	Empty request	Available indicator names
`compute-indicators`	Required `df` and `indicators`; optional `params` and `ticker`. A top-level `result_id` does not replace `df`	Table with `indicator_df` and metadata, inline or stored
`save-dataframe`	Required `dataframe`, `ticker`, `dataset_name`, `start_date`, `end_date`, and `timeframe` (or `interval`); optional `file_format` and `data_kind`	Artifact metadata including `artifact_id`

The same server also provides Shared MCP Tools.

Examples

Retrieve Daily Prices

Call fetch-ohlcvs:

{
  "request": {
    "tickers": ["SPY"],
    "source": "yahoo",
    "timeframe": "1d",
    "start_date": "2026-01-01",
    "end_date": "2026-02-01"
  }
}

Always supply both dates and either timeframe or interval. Their null schema defaults do not make them optional at runtime: omitting the dates or bar size fails before data retrieval. Inspect the actual returned date range, ticker metadata, and row count. Market-data availability and date-boundary behavior depend on the provider.

Calculate an Indicator

Call list-supported-indicators with {"request": {}} to inspect supported names. This small compute-indicators example uses supplied synthetic prices:

{
  "request": {
    "df": {
      "rows": [
        {"timestamp": "2026-01-05", "open": 100, "high": 102, "low": 99, "close": 101, "volume": 1000},
        {"timestamp": "2026-01-06", "open": 101, "high": 103, "low": 100, "close": 102, "volume": 1100},
        {"timestamp": "2026-01-07", "open": 102, "high": 104, "low": 101, "close": 103, "volume": 1200}
      ]
    },
    "indicators": ["SMA"],
    "params": {"SMA": {"timeperiod": 2}}
  }
}

For real data, call result_get on a stored fetch and pass result.payload.ohlcv_df as df. Do not pass a bare result_id, df: {"result_id": "..."}, or the whole response envelope. Indicator warm-up rows may be missing; do not silently treat them as zeros.

Export a Table

Call save-dataframe with the table and required identifiers. A top-level result_id or variable does not replace the required dataframe field. For a stored result, call result_get first and extract result.payload.ohlcv_df after fetching or result.payload.indicator_df after computing indicators. Pass that table as dataframe, not the whole response envelope. This example uses a small synthetic table:

{
  "request": {
    "dataframe": {
      "rows": [
        {"timestamp": "2026-01-05", "close": 101},
        {"timestamp": "2026-01-06", "close": 102}
      ]
    },
    "ticker": "TEST",
    "dataset_name": "daily-prices",
    "timeframe": "1d",
    "start_date": "2026-01-05",
    "end_date": "2026-01-06",
    "file_format": "csv"
  }
}

Read the returned artifact with artifact_read_chunk. CSV and Parquet are suitable exchange formats; pandas pickle should only be used in a trusted environment. Export also requires both dates and a timeframe even though their schema defaults are null. Set them explicitly. Supported file_format values are csv, parquet, and pandas (pickle).

Results and Limitations

Use result_get for a stored output, then inspect result.payload.
When moving data to another MCP server, send the table contents: a result_id only works on the server that produced it.
Preserve timestamps and ticker separation when computing returns or building panels.
A successful connection does not guarantee provider coverage, an active provider subscription, or an unrestricted data quota.
Unsupported indicators, nonnumeric prices, and missing required columns are input errors; provider failures are separate.

Overview

MCP Setup

Fama-French Replicate Server

Statistical Factor Models Server

Jump Models Server

Wavelet Mean Reversion Server

EP Ratio Screener Server

Volatility Scaling Lab Server

Dataloader Server

Overview

Connection

Tools

Examples

Retrieve Daily Prices

Calculate an Indicator

Export a Table

Results and Limitations

​Overview

​Connection

​Tools

​Examples

​Retrieve Daily Prices

​Calculate an Indicator

​Export a Table

​Results and Limitations

Overview

Connection

Tools

Examples

Retrieve Daily Prices

Calculate an Indicator

Export a Table

Results and Limitations