- DISCOVERY
- AUDIT
- VALIDATION
- HUMAN
Short answer
How can I find and audit datasets using my AI?
Your AI uses the data.world catalog to locate datasets and pull their specific metadata or field definitions. You'll get clear answers on what's in a dataset and how it's licensed. Here is how it works.
The results
What your AI delivers
The AI does the heavy lifting of searching and reading, so you can focus on selection.
DISCOVERY
Targeted dataset searches
The AI scans the catalog for specific topics or keywords. You get a list of relevant datasets immediately.
AUDIT
Deep metadata inspection
The AI pulls field definitions and license types. You can verify if a dataset fits your legal or technical needs.
VALIDATION
Data structure checks
The AI looks at the specific fields within a dataset. This helps you confirm the data contains the columns you actually need.
HUMAN
Final data selection
The AI presents the facts, but you decide which datasets are worth downloading or using in your project.
The workflow
What your AI does when the request arrives.
The AI acts as your researcher, navigating the catalog and reading the fine print for you.
Locate data
The AI looks through the data.world catalog to find datasets that match your specific search terms.
search_catalogPull details
Once a dataset is found, the AI grabs the full metadata and license information.
get_dataset_detailsInspect fields
The AI reads the specific field definitions to see exactly what the data contains.
get_dataset_detailsSummarize findings
The AI presents a clean summary of the dataset's contents and usage rights in your chat.
get_dataset_details
Try it
Copy these to start.
Use these prompts to kick off your search or audit.
Starting points
Swap out the bracketed terms for the specific topics or dataset names you're looking for.
data.world Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_tdQsdtjeyEJe7LirxBqlY8zM7HmSW2F7mw85gNx5/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — data.world capabilities are ready to use.
{
"mcpServers": {
"dataworld-mcp": {
"url": "https://edge.vinkius.com/vk_preview_tdQsdtjeyEJe7LirxBqlY8zM7HmSW2F7mw85gNx5/mcp"
}
}
}Search the data.world catalog for datasets related to [climate change trends].
Get the field definitions and license info for the [global weather] dataset.
Find datasets in data.world that contain [retail sales] information.
Audit the [user demographics] dataset and tell me what columns are available.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect data.world once, then ask.
Click the link to connect. Your credentials stay encrypted, and you can start asking questions immediately.
Connect data.world to your AIFAQ
How this task behaves.
- 01
Can the AI download the actual data files?
No. The AI can only read the catalog, metadata, and field definitions. You must download the files yourself through the data.world website.
- 02
Can the AI change the license of a dataset?
No. The AI can only read and report the existing license information.
- 03
Can I search for datasets by specific column names?
Yes, you can ask the AI to find datasets that contain specific fields or columns.
- 04
Does the AI see private datasets?
The AI can only see datasets that your account has permission to access in data.world.
- 05
Can the AI create new datasets in my account?
No. This MCP is for searching and auditing existing catalog information only.
More questions about data.world? See everything the data.world Connector can do
Connect data.world to Claude, Cursor, ChatGPT & more
