Cudf Analytics

Wraps NVIDIA RAPIDS cuDF to run pandas-style data operations on GPU instead of CPU. You get groupby aggregations, statistical summaries, and anomaly detection with the same API you already know, just faster on large datasets. The initialization block is actually useful since it includes a real GPU compute test and graceful pandas fallback, not just import checking. Best for CSV analysis where you're doing repeated aggregations or correlation work on millions of rows. The to_pandas() wrapper with Arrow fallback handles the common gotcha where GPU to host transfers fail. If your dataset fits in memory and you're doing one pass analysis, the speedup might not justify the GPU dependency.

skills.sh installs: 216
GitHub stars: 31.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add langchain-ai/deepagents --skill cudf-analytics --agent claude-code

Installs into .claude/skills of the current project.

Files

cuDF Analytics Skill

GPU-accelerated data analysis using NVIDIA RAPIDS cuDF. cuDF provides a pandas-like API that runs on NVIDIA GPUs, enabling massive speedups on large datasets.

When to Use This Skill

Use this skill when:

Analyzing CSV files, datasets, or tabular data
Computing statistical summaries (mean, median, std, quartiles)
Performing groupby aggregations
Detecting anomalies or outliers in data
Profiling datasets with millions of rows
Computing correlation matrices

Initialization (REQUIRED)

Always start every script with this boilerplate. It tests actual GPU operations, not just import.

import pandas as pd

try:
    import cudf
    # Smoke-test: verify GPU compute AND host transfer both work
    _test = cudf.Series([1, 2, 3])
    assert _test.sum() == 6
    assert _test.to_pandas().tolist() == [1, 2, 3]
    GPU = True
except Exception as e:
    print(f"[GPU] cudf unavailable, falling back to pandas: {e}")
    GPU = False

def read_csv(path):
    return cudf.read_csv(path) if GPU else pd.read_csv(path)

SKILL.md on GitHub