How MCP Connects Cloud AI to ATLAS.ti Projects (Full Transcript)

[00:00:00] Speaker 1: What is MCP? It stands for Model Context Protocol. See it as a bridge that connects AI2, which is cloud AI, to your project in AtlasTI so that you can use AI to help you to analyze your data or do some queries. Let me show you how MCP works. Let's say you connect tools like Calendar, your email system, resources like files, and also AtlasTI project to cloud so that you'll be able to have communications. If your calendar is connected to cloud, you can ask cloud a question about your calendar and information will send through MCP, communicate with your calendar, and then you'll receive that information. So see this information here as context. Without this calendar connected to cloud through MCP, when you ask about what meetings do I have today, cloud might not be able to answer that question. But because you have connected your calendar to cloud, you'll be able to get a very good response. The same thing as file. You can connect your file through MCP and then ask the system questions and the system will be able to access your file and provide you an answer. The same thing as your email, you can do that. So now, if you have access to AtlasTI, the current version, which is Atlas 26.1, you can use cloud to communicate with your files and your quotation codes in cloud and you'll be able to get some answers. Another brief information here, just to help you to understand it better, you have AtlasTI project. So in your project, you have documents, you have codes, you have memos, you have quotations, and also relationship between concepts. Everything is here, right? So with MCP, you are able to connect this project to cloud so that when you ask the system about, okay, can you code my data for me? Or can you extract information that is significant? Cloud will access your project through MCP so that it can provide you an answer that meets your expectations, right? So that's how this is done. Why do you need this one? I think it's a secure way of you using AI to help you to make sense of your data, right? So AI will just securely communicate with AtlasTI through MCP so that you can be able to get information from there. So it makes the data analysis faster. Not only faster, it also helps you to use AI as a collaborator. So you can still analyze your data using AtlasTI, but you can also get ideas and also support from AI so that you'll be able to adequately make sense of your data and also get the findings that will help you to address your research question that you have. So let's go to AtlasTI, and I'm going to show you how I was able to connect my AtlasTI to cloud. So first of all, you have to have the current version of AtlasTI. If you open AtlasTI and you go to options, you click on options, and then you click here, AtlasTI. You see this place, check for updates, right? So the current one you see here, check for update. You can click here and make sure that your AtlasTI is up to date. So after that, what you have to do is to open your project. So I have my project here. I have five documents or five transcripts. I have done my coding based on my research question. I have the codes here, and I don't have any memos. I haven't categorized them to develop themes yet, so I don't have anything here. So the next step is to go to files here. You go to file, and then you click on options, and then make sure that you click on AtlasTI. When you scroll up, this is what you're going to see. Make sure that you check enable MCP server, and then you see here how to cloud desktop, right? You click on that, and then you look for a place to save the file, right? But before you save the file, you create a folder. Based on what you see here, you can easily create a folder. I've already created my folder, right? If you want to create a folder, you go to new folder, right? And then you type mcp atlas.ti, right? So I've created my folder here, and make sure that you know where you created a folder. I think it's created under local disk. I go to users. I go to info. Yours will be very different, maybe a different thing, but the most important thing is that make sure that you know where you save this folder because that's going to be useful when we go to cloud AI. So I double click on that, and then click on select folder, and then click on okay. So this means that the folder has been saved, right? Now we can go back. You have a project here. Make sure that your project is open. Don't close your project, and then we go to cloud AI. Make sure that you have downloaded cloud desktop, right? This is how cloud AI looks like on your desktop. So after downloading cloud desktop, you go here, this corner, click here, and go to file, and you go to settings, and go to extension, and then click on advanced settings. This is what you're going to see when you click on advanced settings. Then click on install extension. Then you look for where you save that information, right? I remember I saved it local drive, users, and info, and document, and then, yeah, this is it. MPC at large. So you just have to know where you save it so that you can look for it, and you can upload it. So you click on preview, and then you click on install, click here, install here. This shows that it has been installed. You can go to configure, click on configure, and close here so that you see the detailed information. So here gives you information about all the tools and resources or files that cloud will be assessing when it asks cloud questions. So you see here it says read only tools. So this means that the system is not going to manipulate any information in your project, right? Now that we are done, we can go back. So click here and go back to the main page. You have to ask cloud to connect to your project, right? So you can say, can you use Atlas TI MCP to explore the code I have developed? So when you're starting the conversation, always mention this one. Can you use Atlas MCP? So I click enter, and let's see what will happen. So this one, you want the system just to explore the codes that you have developed. So it has explored and given you some information. We can do a little comparison and see what the system is saying is true about the number of codes and the kinds of codes that I have. So I think all of the codes are here. It means that the system has access to the project. That's a very good news. And then for the second research question, you also have all these codes here. Thank you so much for your time.