By Kavita Mistry 
Model context protocol, popularly known as MCP, is the next big thing in this AI era. At a high level, the MCP servers allow you to connect your AI chatbot to any apps or services and provide more context for your AI model.
This eliminates the need to switch between apps and manually copy information between your app and AI chatbots like Claude AI, ChatGPT, or even Gemini.
Using these MCP servers, you can connect your favorite apps and services like Slack, Telegram to Claude AI and perform tasks for you.
Best MCP Servers for Claude AI
Gmail MCP Server
Even in 2025, the majority of professionals, even personal conversations, still depend on email, and Gmail remains the popular email client among other email clients, especially for Android users.
Using the Claude Gmail MCP server, you can connect your Gmail app to the ClaudeAI and also analyze all your emails.
And if you are someone who is a professional, sorting, searching, and responding to every email can be overwhelming. This is where you can connect your Gmail account with the Claude AI MCP server and find important messages quickly. Summarize your email content, draft emails, and create summarized responses.
Apart from just reading this, the Gmail MCP server also comes with advanced features, and you can send emails directly from the Claude MCP.
Gmail MCP Service is useful for the majority of users; however, setting up can be overwhelming even for beginners because accessing and authenticating with your Gmail account requires you to create a Google Cloud project, enable Gmail API, and set up Oauth credentials, and later connect it to Cloud MCP.
However, to make it simple, I have provided a simple and easy process to help you get started.
How to connect Gmail to Claude using Gmail MCP Server
- To create a Google Cloud project, first go to the Google Cloud Console dashboard.
- Here, click on the new project, or you can also use an existing Project; however, for this MCP, I would request you to create a new project.
- Now, click on the menu in the top left corner of the screen and select API & Services. Then click on Enabled APIs and Services.
- Now, click the Add API or Enable APIs and Services button at the top.
- And search for Gmail API and enable it
- Now again, click on the menu in the top left corner of the screen and select API & services, and go to credentials
- Now, click on the credentials and select OAuth client ID.
- Now, enter the required details such as app name, type of app, and client name. Once you have successfully created this, Google Cloud Console will generate a JSON file containing your credentials.
- Download the file and rename it to gcp-oauth.keys.json and save it on your device.
- Now you need to install the Gmail MCP server on your device. Open terminal and enter the following command: npx -y @smithery/cli install @gongrzhe/server-gmail-autoauth-mcp– client claude
- And now clone the repository: git clone https://github.com/GongRzhe/Gmail-MCP-Server.git
cd Gmail-MCP-Server - Next, you need to authenticate the server. Open the terminal again and enter the following command: mkdir -p ~/.gmail-mcp
- Next, move the gcp-oauth.keys.json file that you have downloaded to this newly created Gmail MCP directory. You can use this command: mv gcp-oauth.keys.json ~/.gmail-mcp/
- Finally, run this command to authenticate: npx @gongrzhe/server-gmail-autoauth-mcp auth.
If you see any errors, run the following commands to fix them:
mkdir gmail-mcp-test
cd gmail-mcp-test
npm init -y
npm install @gongrzhe/server-gmail-autoauth-mcp
npx @gongrzhe/server-gmail-autoauth-mcp auth
node ./node_modules/@gongrzhe/server-gmail-autoauth-mcp/bin/auth.js
It will open a new tab, and you have to sign in with your Gmail account to authenticate. It generally takes 1-5 hours. If you are facing any errors, try again after some time.
Once you have successfully authenticated, open the Claude desktop app on your device, go to settings, and then developer tab, and click on the edit config file, and add the following code to the edit config file.
{
“mcpServers”: {
“gmail”: {
“command”: “npx”,
“args”: [
“@gongrzhe/server-gmail-autoauth-mcp”
]
}
}
}
And that’s it. Now, restart the device, and that’s it. You have connected your Gmail MCP server to the Claude AI.
Google Drive MCP Server
Google Drive is also one of the most popular cloud storage services used by many people. If you are on Android, it is the default cloud storage.
It offers 15 GB of free file storage, which stores all of your critical data, including data from your Google services, and also WhatsApp data if you have connected and backed it up.
However, if multiple files, apps, and services are connected to your Google Drive, finding the right file you are looking for can be overwhelming in the pool of unknown files.
Thanks to the Claude MVP MCP server, you can connect your Google Drive directly to the Claude AI and let you organize and search the files you are looking for directly from the Claude interface using natural language.
How to connect your Google Drive to your Claude UI using MCP
- Note: For all Google services, you need to get an API through the Google Cloud API platform. You can use an existing project, or you can follow the steps to create a new project and get your new API.
- First, head over to Google Cloud Console and click on Create New Project. You can also select the existing projects if you want.
- Now, click on the menu and select API & Services. Then click on Enable APIs and Services, search for Google Drive, and click Enable.
- Click on the menu, go to the APIs & Services tab, click on Credentials, and then add new credentials. Choose External App, and fill out the required fields such as support email and others.
- Once you have created it, download the JSON file and rename it to gcp-oauth.keys.json and save the file locally on your device.
- Now, use the following code and clone the Google Drive MCP server locally to your device: git clone https://github.com/modelcontextprotocol/servers-archived.git
cd servers-archived/src/gdrive - Next, move the gcp-oauth.keys.json to the gdrive folder that you have created, and next start the server using this: npm start.
Now, add this file to the Claude desktop config file
{
“mcpServers”: {
“gdrive”: {
“command”: “node”,
“args”: [
“/path/to/servers-archived/src/gdrive/index.js”
]
}
}
}
Telegram MCP Server
Telegram is the fourth-largest messaging platform around the world, and it is one of the largest to host groups and channels to communicate with users.
Telegram also supports bots, which help you manage Telegram groups easily. However, connecting it to the LLMs like Claude AI can give extra opportunities like posting content automatically, spotting spam, and also searching chat history and summarizing based on your use case.
Not only this, you can also get a summary of all the chats that are present in the group. You can also delete, create a group, get participant list, list all the admins, get a list of full banned users, and more, the options are endless.
Using it with any LLM, such as Claude AI, has an extra added advantage, like if you are someone who is using Claude to plan your content strategy, you can use the same LLM to post content directly to the Telegram platform.
How to connect your Telegram to the Claude MCP server
- First, you need an API development tool to connect your Claude to your Telegram account. You can visit Telegram.org on your browser and log in to your Telegram account using your phone number.
- Next, click on the API development, and here click on create a new app. Fill in all the required details, such as title, and after Telegram will grant you an API ID and API hash. Keep them safely and securely.
- Next, you need to clone the Telegram repo using the following code: git clone https://github.com/chigwell/telegram-mcp.git
cd telegram-mcp
npm install - Now, enter your API credentials to authenticate: export TELEGRAM_API_ID=your_api_id_here
export TELEGRAM_API_HASH=your_api_hash_here
replace your API here with your API codes that you have obtained from Telegram - Next, you have to set your Telegram API credentials as environmental variables. Follow the code below to get started, and finally, you have to run the server. Use this command to run the server: npm start
- Now, copy and paste the configuration file to your Claude desktop AI, and you have successfully connected your Telegram account to Claude AI. Here you can list all your Telegram chats, send a message, Telegram history, and more.
Now, you need to connect your Claude desktop. Paste the code into the Claude config file
{
“mcpServers”: {
“telegram”: {
“command”: “node”,
“args”: [
“/path/to/telegram-mcp/index.js”
]
}
}
}
Travel Planner MCP Server
Planning travel with the AI is one of the most useful and also frequently used tools; it helps you to save time, compared to opening 10 tabs and analyzing each travel itinerary.
Thanks to AI LLMs like Claude AI, you can get instant personalized travel itineraries. You can also ask questions and modify your travel itinerary anytime. You can also connect apps like To do list using the MCP servers and create an action plan for your travel.
How to connect the Travel Planner MCP server to Claude AI
- First, you need to connect your Claude to the travel planner MCP server. First, you need to ensure that you have installed Node.js, npm, and the Claude desktop app on your device.
- Once you have installed, clone the repo by using the following command: git clone https://github.com/GongRzhe/TRAVEL-PLANNER-MCP-Server.git
cd TRAVEL-PLANNER-MCP-Server - Install all the dependencies with this command: npm install, and start the server using this command: npm start
- Now, add this code to the Claude desktop config file:
{
“mcpServers”: {
“travelplanner”: {
“command”: “node”,
“args”: [
“/path/to/TRAVEL-PLANNER-MCP-Server/index.js”
]
}
}
} - Now, start the MCP server, find your Claude desktop file, and add the travel planner MCP server to the list. Once you have done, save and restart your Claude desktop and you will be able to connect it to the Google Maps, and you can start planning your trips.
Spotify MCP Server
Spotify is the most popular music streaming service around the world. In fact, it is a top music streaming service with more than 35% market share.
With the MCP server, you can connect your Spotify account to the LLM and control your Spotify music hands-free.
While it also comes with start and pause playback controls, search and find music client, get info about the track, and manage Spotify, the biggest advantage of this MCP server is the ability to manage, create, and update your existing playlists.
Managing Spotify playlists is one of the painful jobs while using the app; however, using the LLM makes it easy for you. For instance, you can completely remove the songs or add the songs related to the same genre or artist, which might take some time if you’re doing it manually.
You can also use it to connect and discover, and also related music tracks, and add them to your playlist.
How to connect the Spotify MCP server to the Claude
- Ensure the Node. js and NPM installed on your device, you also need to have a Spotify developer account to connect your Spotify account to the Claude MCP server. A Spotify developer account is free. You can sign up for free on the website.
- Once you have successfully signed up, go to the Spotify developer dashboard and on the top, click on create an app, give it a name, and a description. Now save the app, and you will get a developer and client ID and secret client code.
- Now, click on the app, open its settings, and click on add. Now you need to clone the repo using the following command and install all the dependencies: git clone https://github.com/varunneal/spotify-mcp.git
cd spotify-mcp - Install the dependencies using: npm install
- Now, create a new .env file and save it at the root of this repo.
SPOTIFY_CLIENT_ID=your_client_id_here
SPOTIFY_CLIENT_SECRET=your_client_secret_here
SPOTIFY_REDIRECT_URI=http://localhost:3000/callback
PORT=3000
Replace these with the Spotify values. - Now, restart the MCP server using this command: npm start
- Now you have to make changes to the Claude config file and add the Spotify MCP server:
{
“mcpServers”: {
“spotify”: {
“command”: “node”,
“args”: [
“/path/to/spotify-mcp/index.js”
]
}
}
} - Now you can play all your Spotify playlist songs and also customize them based on your preferences.
Terminal Controller
Using the command line or a Terminal can be overwhelming, especially if you are a beginner.
Still, the majority of applications require terminal knowledge to make the apps work. For instance, connecting your Claude MCP Service to the apps also requires Terminal knowledge.
Fortunately, thanks to the terminal MCP server. Now, you can connect the Claude MCP server and use a natural language to connect and write commands in the terminal.
And also, not everyone can remember terminal commands. If you forgot a command or are having a hard time interacting with a terminal, you can simply use natural language, just like how you chat and interact with the AI chatbot.
If you are worried about making unwanted changes. Don’t worry. The Claude MCP also comes with an extra layer of safety, by telling you what the command will do. You can confirm it before taking any risks and running the command.
How to connect your Claude MCP server to your terminal
- To use the terminal first, you need to ensure you have Node.js installed on your device, and also npm, and you have successfully installed the Claude desktop server and are running it on your device.
- Once you have the device, you have to clone and install the terminal MCP server on your device. First, use the following command to install it on your device, and also install the dependencies by following the command below, and then start running the server.
git clone https://github.com/GongRzhe/terminal-controller-mcp.git
cd terminal-controller-mcp - Now run the server using this command: npm install, and then start the server: npm start
- Next, you need to connect your Claude MCP server to the Claude desktop, open your Claude desktop settings file, and add the server to it.
{
“mcpServers”: {
“terminal”: {
“command”: “node”,
“args”: [
“/path/to/terminal-controller-mcp/index.js”
]
}
}
} - Once you have configured the Claude config file, you can use it to list out all the commands, such as restart, all files in your specific directory, system error, delete the file, and more. You can interact with Claude directly using natural language, and your terminal does the job for you.
Todoist
If you have been following us, you know how much we love to-do lists. We have plenty of how-to guides covering using the best techniques, and guides on creating a personalized to-do list app.
Fortunately, you can now directly connect your Claude or any other LLM to your AI and manage all your tasks in one place.
Since many LLMs have also now started rolling out a memory feature, which can remember all the details that you have interacted with the chat. Using this feature, you can plan your day using the LLM.
You can directly use a Claude LLM to quickly add updates and organize tasks without remembering them yourself.
It not only makes it work for you easier but also helps you to organize tasks and create tasks that are more action-based instead of creating vague tasks, plus you can also use several to-do list principles in the LLM itself. Once you have a detailed breakdown of your task, you can add it to your to-do list since it is the most popular and helpful.
We have a detailed guide on how to create personalized and actionable to-do list tasks. You can use those principles to create better tasks.
How to connect Claude to the Todoist app
- Before connecting your Claude AI to your Todoist, you need to have an account in Todoist.
- If you don’t have an account, create an account, which is a must to connect your work with Claude AI LLM. If you have been using it, you can also use the TickTick to-do list app. If you don’t want the Todoist app, the process is very similar. If you are using the TickTick app, go to the website and create an account.
- Now, once you log in with your account, go to settings and integration, and find the API token and copy it.
- Now, use the following command to clone the repo and dependencies.
git clone https://github.com/abhiz123/todoist-mcp-server.git
cd todoist-mcp-server - Now, install all dependencies using this command: npm install.
- Create a .env file in the root of the project, and here replace the actual API token and add it to the required code in the following command.
TODOIST_API_TOKEN=your_todoist_api_token_here
PORT=3000 - Now, start the server: npm start
- Finally, connect the desktop to other MCP clients and add the MCP client to your Claude desktop AI.
{
“mcpServers”: {
“todoist”: {
“command”: “node”,
“args”: [
“/path/to/todoist-mcp-server/index.js”
]
}
}
} - Once you have successfully edited the file and added it to your Claude desktop, here you can create your new card, show all due today, mark your task complete, and list the task in one place.
WhatsApp is also one of the most popular messaging platforms around the world. In India, it is a default messaging platform for the majority of users.
With the introduction of channels and groups, WhatsApp has also become a primary communication platform for many teams and plenty of communities. Now with the Claude MCP server, you can directly connect your WhatsApp account to the Claude AI and get access to all your messages, summarize them, and also send messages to your WhatsApp directly from the Claude interface.
Thanks to the Claude MCP server, you can also bring features that are not yet available on WhatsApp. For instance, you can also schedule messages and even summarize all your group conversations in a single place from your Claude interface.
Plus, if you are using a team, you can also summarize the messages and connect them to your to-do list app and add them to your tasks, which makes it overall the best possibility to connect your WhatsApp account to the Claude MCP desktop server.
How to connect your Claude MCP server to the WhatsApp MCP server
- To set up the WhatsApp MCP server and connect to Claude, you need to install Node. JS on your device and npm desktop. \
- Now, ensure you have a WhatsApp active account running on your phone and install the Google Chrome browser for WhatsApp web automation.
- Once you have done this, clone the repository using the command below and install all the dependencies using the npm install command.
git clone https://github.com/lharries/whatsapp-mcp.git
cd whatsapp-mcp
{
“mcpServers”: {
“whatsapp”: {
“command”: “node”,
“args”: [
“/path/to/whatsapp-mcp/index.js”
]
}
}
} - Next, install dependencies: npm install, and start the MCP server: npm start
- Now, start the MCP server and finally connect your Claude configuration account and start using your WhatsApp.
Phone MCP Server
Phone MCP server is the most powerful MCP server, which lets you automate and control your Android smartphone.
Using this MCP server, you can directly connect your Android smartphone to the Claude MCP server, and you can make a call, send text messages, play music, summarize your notifications, and everything else using your natural language by using the Claude desktop interface.
You can completely interact with your smartphone and all the functions from the Claude desktop interface.
How to connect your Android smartphone to the Claude MCP server
- To connect your smartphone to the Claude MCP server, first, you need to install Node. js, npm, and also ADB on your device. Follow this video to install and set up ADB on your device.https://youtu.be/GERlhgCcoBc
- Once you have installed ADB, ensure your smartphone has developer mode and also USB debugging enabled.
- Once you have done, install Phone MCP server using the command below and install all the dependencies.
git clone https://github.com/hao-cyber/phone-mcp.git
cd phone-mcp - Ensure you have USB debugging enabled on your device. You can test it with this command in your terminal: adb devices
- Install npm install abd run the server using npm start
- Connect your mobile to a USB and authorize the MCP server, edit the complete file in the Claude desktop app, and successfully connect your smartphone to your Claude MCP server.
{
“mcpServers”: {
“phone”: {
“command”: “node”,
“args”: [
“/path/to/phone-mcp/index.js”
]
}
}
}
Microsoft PowerPoint MCP server
Finally, ending the list to automate the most boring tasks in your day-to-day workflow.
The Microsoft PowerPoint MCP server lets you create, edit, and automate presentations all using the natural language itself.
Since you are directly writing with Claude, you can also automate this process by adding the relevant information, relevant sources, and everything. Not only are you automating the current PowerPoint creation process, but you are also adding more context and creating high-quality PowerPoint presentations, which can be especially useful for students or someone who is a working professional to create detailed and advanced presentations.
How to connect your Claude to your Microsoft PowerPoint presentations
- First, install Node. JS, NPM, and Microsoft PowerPoint on your device.
- Now, clone the repository using the following terminal command and install all the dependencies by using the npm install command.
git clone https://github.com/GongRzhe/Office-PowerPoint-MCP-Server.git
cd Office-PowerPoint-MCP-Server - Install dependencies using this command: npm install abd start the server: npm start
- Now, run the MCP server using npm start, and finally connect the PowerPoint MCP server to the client setup.
{
“mcpServers”: {
“powerpoint”: {
“command”: “node”,
“args”: [
“/path/to/Office-PowerPoint-MCP-Server/index.js”
]
}
}
}
And that said, you have successfully connected your Microsoft PowerPoint to the Claude MCP. You can directly interact with it and start creating your presentations. You can create a new presentation, add a slide, set an image, export, and also summarize your current slides and more.
Connect Your Favorite Apps to Claude AI
We are just getting started. Fortunately, there are plenty of MCP servers available for Claude AI already, which allow you to interact with your favorite apps.
For instance, the recent Zerodha Kite MCP lets you connect your Kite trading account to Claude AI and analyze your stock portfolio, keep track of your stocks, and get detailed reports.
Also, our first Windows File MCP server allows Claude to connect with a local file system on your device and interact with your files.
The best part is, MCP services are free to get started with your Claude AI, and you can automate your daily boring tasks, and also get detailed analytics of your current process.
There are hundreds of MCP servers that are being built every single day that can help any user connect and interact with their favorite apps and services. If you are having any issues while connecting these MCP servers. You can visit the details for MCP instructions, and also comment below, if you have any issues and errors while connecting to these MCP services.