Getting started with the Dataprovider MCP: let AI query 350 million domains for you
AI assistants are powerful, but they're only as useful as the data they can access. By default, tools like Claude and ChatGPT work from their training data, which means they can't tell you how many Shopify stores launched in Germany last month or which payment provider is gaining market share in Scandinavia.
The Dataprovider MCP changes that. It connects AI assistants directly to our database of 350+ million domains and 200+ structured data fields, giving you real-time answers to questions that would otherwise require hours of manual research.
What is MCP?
MCP stands for Model Context Protocol, an open standard developed by Anthropic that lets AI assistants connect to external data sources and tools. Instead of copying data into your prompts, MCP enables the AI to query Dataprovider directly during your conversation.
Think of it as giving your AI assistant a direct line to our search engine. You ask a question in natural language, the AI figures out which filters and fields to use, runs the query against our database, and returns structured results — all within the same conversation.
The Dataprovider MCP works with Claude (Anthropic), ChatGPT (OpenAI), and any other MCP-compatible client. No coding required.
What you can do with it
The MCP gives your AI assistant access to the same data and tools available in our search engine, but through natural conversation.
You can:
- Search and filter websites. Ask for all eCommerce stores in France using Shopify, or find businesses in a specific city that don't have an SSL certificate. The AI translates your question into the right filters automatically.
- Analyze trends over time. Track how technology adoption changes month by month. See whether WordPress is growing or shrinking in a specific market, or how a competitor's traffic has developed over the past year.
- Look up any domain. Enter a hostname and get a full profile: company information, technologies used, traffic metrics, hosting details, security scores and more.
- Compare and aggregate. Get breakdowns of CMS market share by country, see which payment providers dominate in a region, or find out how many websites were added to a specific TLD last month.
- Discover connections. Find all websites owned by the same entity, explore subdomains, or identify sites that are similar to a given domain based on technology stack and content patterns.
- Run reverse DNS lookups. Enter an IP address and see which domains are hosted there — useful for infrastructure research and security analysis.
All of this happens in plain language. You don't need to know our query syntax or field names. The AI handles that translation for you.
How to set it up
Getting started takes about two minutes. The process is the same for Claude and ChatGPT: you create an access token in your Dataprovider account and add the MCP URL to your AI assistant.
Step 1: Create your access token. Log in to your Dataprovider account and navigate to the MCP section. Click "Create MCP access token" to generate your personal key. Keep this token safe — you'll need it in the next step.
Step 2: Add the connector. In Claude, go to Admin Settings, then Connectors, and click "Add custom connector." In ChatGPT, the process is similar through the settings menu. Enter a name (for example "Dataprovider") and paste the MCP URL with your token:
https://mcp.dataprovider.com/mcp?api-key=YOUR_ACCESS_TOKEN
Step 3: Enable it. Toggle the Dataprovider connector on, and you're ready to go. The AI assistant now has access to all available tools and data fields in your subscription.
For detailed step-by-step instructions with screenshots, see our setup guides:
- Add Dataprovider MCP to Claude
- Add Dataprovider MCP to ChatGPT
Custom implementations: If you're building your own integration or using a different MCP-compatible client, use the server URL directly: https://mcp.dataprovider.com/mcp?api-key=YOUR_ACCESS_TOKEN
Example queries to try
Once connected, just start asking questions. Here are some examples organized from simple to advanced.
Basic lookups
- "What technologies does nike.com use?"
- "Show me the traffic trend for shopify.com over the past 12 months"
- "Who owns the same domains as booking.com?"
Market research
- "How many eCommerce stores are there in the Netherlands?"
- "What are the top 10 CMS platforms used in Germany?"
- "Show me the market share of payment service providers in Scandinavia"
Competitive intelligence
- "Find websites similar to hubspot.com"
- "Which analytics tools are most popular among Shopify stores in the UK?"
- "Compare WordPress vs Wix adoption trends in France over the past 2 years"
Lead generation
- "Find eCommerce stores in Belgium that use Magento but don't use Google Analytics"
- "Show me businesses in California with a trust grade of A and more than 10,000 monthly traffic"
- "List all websites in the Netherlands that added Shopify last month"
Brand protection
- "Find domains that are cybersquatting targets of amazon.com"
- "Show me recently registered domains that contain the brand name 'adidas'"
- "Which websites mention the brand 'Rolex' but are hosted in countries known for counterfeits?"
- "Find eCommerce stores with a low trust grade that sell products mentioning 'Nike'"
Security and infrastructure
- "Which high-traffic websites in Germany don't have a valid SSL certificate?"
- "Do a reverse DNS lookup for IP address 104.21.32.1"
- "Find domains with an expiring SSL certificate in the next 30 days"
Tip: You don't have to be precise. The AI understands context and will figure out the right fields and filters. If the query is ambiguous, it will ask you to clarify. You can also follow up on any result: "Now filter those for only B2B companies" or "Show me the trend for the top 3."
Tips for getting the best results
Start broad, then narrow down. Begin with a general question and refine from there. "How many Shopify stores are in Europe?" followed by "Now break that down by country" works better than trying to get everything in one query.
Ask follow-up questions. The AI remembers context within your conversation. After a lookup, you can say "Now show me the trend for that" or "Filter those for eCommerce only" without repeating yourself.
Be specific about what you want to see. If you need certain fields in the output, just say so. "Show me hostname, company name, country and CMS" gives you a cleaner result than a broad "show me everything."
Use it for exploration. The MCP is not just for answering questions you already have. Try things like "What's interesting about the .ai TLD?" or "What trends do you see in the Dutch eCommerce market?" and let the AI guide the analysis.
Know your subscription scope. The fields and data available through the MCP match your Dataprovider subscription. If a query returns no results for a specific field, it may not be included in your plan.
Getting help
If you run into issues during setup or have questions about what's possible, reach out to our team through the contact page or check the MCP section in your Dataprovider console.
The MCP is the fastest way to get from a question to an answer. No dashboards to configure, no exports to download, no queries to write. Just ask.