# jdatamunch-mcp

Source: https://glimind.com/server/pypi%2Fjdatamunch-mcp
Registry id: `pypi/jdatamunch-mcp`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "jdatamunch-mcp": {
      "command": "uvx",
      "args": [
        "jdatamunch-mcp"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add jdatamunch-mcp -- uvx jdatamunch-mcp
```

## Tools (23)

- `aggregate`: Perform server-side aggregation with group_by
- `check_column_drop_safe`: Check if a column can be safely dropped
- `data_health_radar`: Return a radar chart of data health metrics
- `describe_column`: Return distribution and statistics for a single column
- `describe_dataset`: Return column names, types, cardinality, null rates, and sample values for a dataset
- `find_similar_columns`: Find columns with similar names or content across datasets
- `find_unused_columns`: Find columns that are rarely used or have low information
- `get_correlations`: Return correlations between columns
- `get_data_hotspots`: Return data quality hotspots (nulls, cardinality anomalies, outliers)
- `get_dataset_health`: Return data quality metrics for a dataset
- `get_distribution`: Return value distribution for a column
- `get_rows`: Return rows matching filters
- `get_schema_drift`: Return schema drift between versions of a dataset
- `get_schema_impact`: Return impact of schema changes on queries and joins
- `index_local`: Index a local dataset file (CSV, Excel, Parquet, JSONL) and return dataset metadata
- `index_repo`: Index datasets from a GitHub repository
- `join_datasets`: Join two datasets and return results
- `plan_query`: Preview cost and result size of a query before running
- `run_sql`: Run a SQL query against the dataset
- `sample_rows`: Return a sample of rows from a dataset
- `search_data`: Semantic search across dataset values
- `suggest_joins`: Suggest possible joins between datasets
- `suggest_keys`: Suggest key columns for joining datasets

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.