CSV Analyzer Pack

Analyze, filter, and transform CSV data with pandas-powered tools. Group, aggregate, merge, sort, and export — all without writing pandas code yourself.

Quick Start

agentnode install csv-analyzer-pack

from csv_analyzer_pack.tool import run

result = run(
    capability="analyze_csv",
    params={
        "file_path": "/tmp/data.csv",
        "operations": [{"type": "describe"}]
    }
)
print(result["summary"])

Usage

Summary statistics

result = run(
    capability="analyze_csv",
    params={
        "file_path": "/tmp/sales.csv",
        "operations": [{"type": "describe"}]
    }
)
for col, stats in result["summary"].items():
    print(f"{col}: mean={stats['mean']:.2f}, min={stats['min']}, max={stats['max']}")

Group and aggregate

result = run(
    capability="analyze_csv",
    params={
        "file_path": "/tmp/orders.csv",
        "operations": [
            {"type": "group_by", "column": "category"},
            {"type": "aggregate", "column": "total", "functions": ["sum", "mean"]}
        ]
    }
)

Filter, sort, and export

result = run(
    capability="analyze_csv",
    params={
        "file_path": "/tmp/users.csv",
        "operations": [
            {"type": "filter", "column": "active", "operator": "==", "value": True},
            {"type": "sort", "by": "created_at", "ascending": False},
            {"type": "select_columns", "columns": ["name", "email", "plan"]}
        ],
        "output_path": "/tmp/active_users.csv"
    }
)
print(f"Exported {result['rows_output']} rows")

API Reference

Capability	Description
analyze_csv	Apply operations to CSV data and return results

Operations: describe, filter, sort, group_by, aggregate, select_columns, merge, deduplicate, fill_missing.

Parameters: file_path (required), operations (required), output_path, encoding.

Returns: data, summary, rows_processed, rows_output, output_path, columns.

Requirements

No environment variables required.

License

MIT