VERIFIED

Kaggle

Safety: 65/100

0.0(0)

This server provides tools to search and download datasets from Kaggle, and generate prompts for creating EDA notebooks, streamlining data exploration workflows.

MCP ServersData & DatabasesSearch & WebProductivityDevelopmentAI & LLMUtilitiesFreemium

Boost this tool

Subscribe to listing upgrades or segmented pushes.

Overview

This server provides tools to search and download datasets from Kaggle, and generate prompts for creating EDA notebooks, streamlining data exploration workflows.

This server is relatively safe for data exploration. The primary risk is related to managing the Kaggle API key securely and ensuring the integrity of downloaded datasets. Avoid downloading datasets from unknown or untrusted sources. The EDA notebook generation is safe as it only generates a prompt, not executable code.

Performance is limited by Kaggle API rate limits and network bandwidth. Downloading large datasets can be time-consuming.

Kaggle API usage is generally free, but excessive usage may be subject to rate limits or restrictions. Consider the cost of storage for downloaded datasets.

Python33 starsMIT

Connections & Capabilities

Connects To

Docker

Capabilities

ReadWrite

Client Compatibility

Claude Desktop

full

Cursor

untested

Windsurf

untested

VS Code (Copilot)

untested

Quickstart

Install

pip install -r

Claude Desktop Config

{
  "mcpServers": {
    "kaggle-mcp": {
      "command": "kaggle-mcp",
      "cwd": "<path-to-their-cloned-repo>/kaggle-mcp"
    }
  }
}

Environment Variables

KAGGLE_USERNAMEKAGGLE_KEY

Tools (3)

safe

search_kaggle_datasets

Searches Kaggle for datasets matching a given query and returns a list of the top 10 results.

Read-only operation; no data modification or system access.

moderate

download_kaggle_dataset

Downloads a specified dataset from Kaggle to a local directory.

Writes files to disk, but within a controlled directory.

safe

generate_eda_notebook

Generates a prompt for creating an EDA notebook for a given Kaggle dataset.

Generates a prompt message; no execution or data modification.

Security

Auth Method

API Key

Known Risks

!Compromised Kaggle API key could lead to unauthorized data access.
!Downloaded datasets may contain malicious code or biased data.
!Insecure storage of the Kaggle API key in `.env` file.
!Potential for denial-of-service attacks by excessive API calls.

Safety Score

moderate risk

Positive

+Requires Kaggle API key, limiting unauthorized access.
+Datasets are downloaded to a specific directory, preventing accidental system-wide modifications.
+Primarily read operations (search, download) with limited write capabilities.
+No direct execution of arbitrary code.

Risks

-Kaggle API key must be stored securely to prevent unauthorized access.
-Downloading datasets from untrusted sources could introduce malicious files.
-The `download_kaggle_dataset` tool writes files to disk.
-EDA notebook generation relies on external AI models, which could introduce biases or vulnerabilities.

Use Cases

Data Exploration

-Searching for relevant datasets on Kaggle.
-Downloading datasets for local analysis.
-Generating EDA notebook prompts to accelerate initial analysis.

Machine Learning Projects

-Finding datasets for training machine learning models.
-Quickly accessing and preparing data for model development.
-Automating the initial EDA process.

Educational Purposes

-Providing students with easy access to datasets for learning.
-Simplifying the process of setting up data analysis environments.
-Generating EDA prompts to guide students in data exploration.

Who Is This For?

Best For

Data scientists exploring Kaggle datasets.
Machine learning engineers prototyping new models.
Students learning data analysis and machine learning.
Researchers seeking publicly available datasets.

Not Ideal For

Applications requiring real-time data access.
Projects with strict data security requirements.
Environments with limited network connectivity.
Automated systems requiring high throughput data processing.

Autonomy & Execution

Default ModeRead and Write enabled

Sandboxed

The server operates within the constraints of the Kaggle API and the local file system. Autonomy is limited by the need for a valid API key and the controlled download directory.

StatelessYes

Retry LogicNot documented

ConcurrencySingle-threaded

Production Tip

Monitor API usage to avoid rate limits and implement robust error handling for production deployments.

FAQ

How do I obtain a Kaggle API key?

Go to your Kaggle account settings and create a new API token. This will download a `kaggle.json` file containing your username and key.

Where should I store my Kaggle API key?

It is recommended to store your API key in a `.env` file and avoid committing it to version control.

What happens if I exceed the Kaggle API rate limit?

The server will return an error message. You should implement retry logic with exponential backoff to handle rate limits gracefully.

Can I download datasets to a specific directory?

Yes, you can specify the `download_path` parameter when calling the `download_kaggle_dataset` tool.

What format does the `generate_eda_notebook` tool return?

It returns a structured prompt message suitable for an AI model to generate EDA code.

Is it safe to download datasets from Kaggle?

While Kaggle datasets are generally safe, it's always a good practice to scan downloaded files for potential malware before using them.

How do I configure the server to use a proxy?

You can set the `HTTP_PROXY` and `HTTPS_PROXY` environment variables to configure the server to use a proxy.

Key Features

Data & Databases Search & Web Productivity Development AI & LLM Utilities

Metrics

Discovered2/14/2026

Reviews0

Saved By0 users

Related Tools

Kaggle | AI Builder Hub Review & Alternatives | Flaex AI