- Home
- Skills
- Data & Databases
- Kaggle Dataset Assistant
Kaggle Dataset Assistant
Helps your AI agent search, inspect and download Kaggle datasets in plain English, using your own Kaggle token.
Free
Works with the AI Agents you already use
Kaggle Dataset Assistant
Example session with this skill installed
Find me a dataset about global CO2 emissions and show me what files are in the best one.
- Read your context and instructions
- Compiled the kaggle dataset assistant
- Generated the data export
I have searched Kaggle for global CO2 emissions datasets. The most relevant result is krupadharam/global-co2-emissions-dataset.
Kaggle Dataset Search Results
| Dataset Reference | Title | Relevance |
|---|---|---|
krupadharam/global-co2-emissions-dataset | Global CO2 Emissions Dataset | Comprehensive historical data by country |
unsdsn/world-happiness | World Happiness Report | Includes environmental factors and emissions |
yoannboyere/co2-ghg-emissionsdata | CO2 and GHG Emissions Data | Detailed greenhouse gas breakdowns |
I have inspected the files inside krupadharam/global-co2-emissions-dataset:
global_co2_emissions.csv(1.2 MB)metadata.json(4 KB)README.md(2 KB)
Next steps
- Would you like me to download and unzip this dataset for you?
- Once downloaded, should I load the CSV to show you the first 10 rows?
- Or should I try a different search query for more specific data?
global_co2_emissions.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Working with raw data usually involves manual downloads, unzipping files, and navigating messy directory structures. This skill automates the data acquisition phase by connecting your agent directly to Kaggle via a Model Context Protocol (MCP) server.
What it does
- Keyword searching locates relevant datasets using natural language queries without leaving the chat interface.
- Remote inspection lists files and extensions inside a dataset before you commit to a download.
- Automated retrieval downloads and unzips datasets into organized subfolders automatically.
- Troubleshooting support identifies specific Kaggle API errors like 401 unauthorized or 403 forbidden issues.
How it works
- Search for datasets using 2 to 4 keywords to find the most relevant
owner/dataset-namereferences. - Inspect the file manifest of a selected dataset to ensure it contains the required CSV or JSON files.
- Download the verified dataset, which the agent unzips into a dedicated local directory.
- Preview the resulting files by loading the first few rows of data to begin analysis immediately.
Frameworks & tools
This skill requires a Kaggle MCP server and a valid Kaggle API token. It works within MCP-compliant environments like Claude Desktop or Cursor.
Why this beats prompting it yourself
Generic prompts often struggle with Kaggle's specific owner/slug naming conventions and the multi-step verification process required to avoid large, useless downloads. This skill enforces a strict "inspect before download" workflow that saves disk space and ensures the agent handles local file paths and unzipping logic correctly every time.
Use cases
- Find and download specific CSV files for machine learning training sets.
- Source public datasets for academic assignments or data science portfolios.
- Inspect large Kaggle repositories to find specific file types before downloading.
- Troubleshoot local Kaggle API credential and configuration issues.
Known limitations
Requires an active Kaggle MCP server installation. Users must provide their own Kaggle API token via local environment variables or MCP config files.
Requirements
This skill does not include the server. You need to install the Kaggle MCP server separately and use your own free Kaggle API token. Setup steps are included inside the skill.
Requires local Kaggle MCP server setup. Users must provide their own API credentials. Does not support datasets requiring manual competition entry or private data access via the API.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 16 days ago
- Free with an Agensi account
Requires