mcp-dataframe-qa

sindhug/mcp-dataframe-qa

★ 0 stars Python Code/Dev Tools Updated 4d ago

Research-informed MCP server for safe dataframe question answering over local data

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "mcp-dataframe-qa": {
      "command": "uvx",
      "args": [
        "mcp-dataframe-qa"
      ]
    }
  }
}

Or install with pip: pip install mcp-dataframe-qa

README Excerpt

A research-informed MCP server for safe dataframe question answering over local data. Dumping an entire dataframe into an LLM prompt for every question is an inefficient way to do analytics. It burns context window on raw rows, gets expensive quickly, forces aggressive truncation for real datasets, and can still leave the model guessing instead of calculating. The better pattern is to give the model compact schema context, let it decide what statistics are needed, execute those specific datafram

Tools (1)

max_execution_ms

mcp-dataframe-qa

Quick Install

README Excerpt

Tools (1)

Related Code/Dev Tools Servers