Would you like to query our entire database and get results instantly, without ever needing to learn how to write a single line of Python code or set up a data API?
Model Context Protocol (MCP) enables you to ask questions and receive answers instantly using AI assistants.
Think of MCP as an interpreter who can take your request, translate it into a query, send a request to our database, and then share the results with you.
You can use this tool to retrieve data from our Multi-Source Company, Multi-Source Employee, and Base Jobs databases.
Ready to start the setup?
Step 1. Install Node.js
To connect to the remote Coresignal MCP server, you must install Node.js on your MacBook.
Node.js is a runtime environment that enables you to run JavaScript code outside of a web browser, on your computer, in a terminal, or within applications. Developers widely use it to build tools, run scripts, and manage servers.
Here is how you do it:
- Go to the Node.js website.
- Download the application.
- Run the program and follow the installation instructions.
Why do you need to install it? To connect to Coresignal’s MCP server, you’ll use a command-line tool called npx. This tool is included with Node.js. Installing Node.js ensures you can run the command needed to securely connect to the MCP server and start accessing Coresignal’s APIs directly from your application or development environment.
Step 2. Install an AI assistant
MCP is a tool that enables AI assistants, such as Cursor, to connect directly to Coresignal’s data. With it, you can easily query the database using natural, human language – the assistants will translate.
For this tutorial, we’ll use Cursor, which is a tools developers typically use, but generally, any large language model would work. This particular tool offers a two-week free trial and has fairly generous request limits, which is great for testing.
While you are installing Cursor, you might get this screen:
If that’s the case, move the Cursor icon into the Applications folder.
You are all set!
Step 3: Create or log in to your free Coresignal account
Ensure you have an active Coresignal account. If you don’t yet have one yet, sign up for a free trial to receive 400 free search credits, which will be enough to start exploring data and testing queries.
What you'll get:
- Access to millions of company profiles, including key industries such as biotech or AI.
- Up to 200 company profiles to download, free of charge.
- The ability to generate up to 400 search queries at no cost.
- Instant access with your Google Business account – no credit card required.
- 14-day free trial included for a deeper look at the data.
Whether you're testing our data for the first time or validating its fit for your use case, you can explore freely before making any decisions.
Start a 14-day free trial (no credit card required)
Instantly access millions of company, employee, and jobs records
Step 4: Retrieve the API key from Coresignal's self-service
Go to your Coresignal's self-service platform, and on the homepage, you will find the API key and the MCP integration code.
Your MCP integration code will have the API key automatically included, so choose the Cursor option and copy the code:
Step 5: Paste the code with the API key
Press the application icon to launch it. Now, click the cog icon in the top-right corner in the Cursor app. It will open the settings.
In the Settings, find the MCP menu item. Click the Add new global MCP server button.
In the Command line, paste your code with the API key. Your final result will look like this:
Do not forget to refresh the tool after the integration. All done!
Step 6: Ask your question
On the right, nearby the Settings icon, you will see three panels that . Choose the Toggle AI Pane button and it will open a panel where you can enter your request.
Let's say you want to find AI companies that have secured funding in 2025. Simply type in that request into the field. If the request is not clear enough, the tool will ask you additional questions to refine it, and will allow you to review the query before it runs.
Important! Instead of Auto-select, choose a specific AI agent. For this tutorial, I'll pick the Claude 3.5-sonnet model.
Step 7: Review the query
Once you enter the request, the AI agent will work as an interpreter that reviews your request, identifies the right API and the specific data points that would work the best, and provides you with a query. You can review the query before running it.
Each request will use one Search query, and each result will use up one Collect query.
Step 8: Check the results
Once you run the query, it will return the results for you to review. You can adjust your request or ask to return specific data points or more information about each company.
Query data more easily starting today
Accessing Coresignal data just got easier, and you can quickly assess it without any technical knowledge. All you need is to set up the MCP access and start querying.
If you have any questions about the data – my team is ready to help! Contact us or reach out to me on LinkedIn.
Start a 14-day free trial (no credit card required)
Instantly access millions of company, employee, and jobs records
