Image by Author
I just discovered the hype around MCP (Model Context Protocol) servers and decided to experience it for myself to see how amazing they really are. I tried the Spotify MCP server and used it with Claude Desktop—and it completely blew my mind! I am so obsessed now that I have started exploring other MCP servers.
So, here is my list of top 10 MCP servers that anyone can use for work, personal projects, or just to marvel at how far AI has come.
1. Filesystem MCP Server
The Filesystem MCP Server enables AI models to seamlessly interact with your local file system, unlocking powerful capabilities for managing files and directories. Here are its key features:
- Read/Write Files: Access and modify files directly.
- Create/List/Delete Directories: Manage folders with ease.
- Move Files/Directories: Organize your file system efficiently.
- Search Files: Quickly locate specific files.
- Get File Metadata: Retrieve detailed information about files, such as size, type, and creation date.
Link to the repository: servers/src/filesystem
2. Google Drive MCP Server
The Google Drive MCP Server integrates with Google Drive, allowing AI models to interact with your cloud storage for efficient file management. Here are its key features:
- Search Files: Search for files in Google Drive using a query string. The server returns file names and their MIME types.
- Read Files: Access and retrieve files directly from Google Drive using their unique file IDs (gdrive:///<file_id>).
- Supports All File Types: The server works with all file types stored in Google Drive.
Link to the repository: servers/src/gdrive
3. Slack MCP Server
The Slack MCP Server integrates with the Slack API, enabling AI models like Claude to interact with Slack workspaces. Here are its key features:
- List Channels: Retrieve a list of public channels in the workspace with their IDs and details.
- Post Messages: Send messages to specific Slack channels.
- Reply to Threads: Add replies to specific message threads.
- Add Reactions: React to messages with emojis.
- Get Channel History: Retrieve recent messages from a channel.
- Get Thread Replies: Fetch all replies in a specific message thread.
- List Users: Retrieve a list of workspace users with basic profile information.
- Get User Profile: Access detailed profile information for a specific user.
Link to the repository: servers/src/slack
4. Memory MCP Server
The Memory MCP server is designed to store and retrieve contextual data for AI models. It acts as a short-term or long-term memory bank, enabling models to recall previous interactions or maintain context across sessions. This is particularly useful for conversational AI applications or task continuity. Here are its key features:
- Entities: Represent nodes in the graph with a unique name, type (e.g., “person”, “organization”), and observations (e.g., “Speaks fluent Spanish”).
- Relations: Define connections between entities (e.g., “John_Smith works_at Anthropic”).
- Observations: Store discrete facts about entities, such as “Graduated in 2019” or “Prefers morning meetings.”
Link to the repository: servers/src/memory
5. Spotify MCP Server
For music lovers and developers alike, the Spotify MCP server allows AI models to interact with Spotify’s API. It can create playlists, recommend songs, or even analyze your listening habits. Here are its key features:
- Control Playback: Start, pause, and skip tracks.
- Search Content: Find tracks, albums, artists, and playlists.
- Get Information: Retrieve details about tracks, albums, artists, or playlists.
- Manage Queue: Add and organize tracks in the Spotify queue.
Link to the repository: varunneal/spotify-mcp
6. Notion MCP Server
The Notion MCP Server integrates with Notion’s API to manage a minimalist personal to-do list through Claude. Here are its key features:
- Add New Tasks: Create new to-do items in the Notion database.
- View All Tasks: Retrieve the full list of to-do items.
- View Today’s Tasks: Filter and display tasks marked for “today.”
- Mark Tasks as Complete: Check off tasks as done.
Link to the repository: danhilse/notion_mcp
7. Email MCP Server
The Email MCP Server enables Claude to handle email functionality, including composing, sending, and managing attachments. Here are its key features:
- Send Emails: Compose and send emails to multiple recipients with optional attachments.
- Search Attachments: Locate files in directories using pattern matching.
- Secure Transmission: Uses SMTP for secure email delivery.
Link to the repository: Shy2593666979/mcp-server-email
8. Windows Control MCP Server
The Windows Control MCP Server enables Claude to programmatically control system operations on Windows, including mouse, keyboard, window management, and screen capture. Here are its key features:
- Window Management: List, focus, resize, reposition windows, and retrieve window details.
- Mouse Control: Move the mouse, click, scroll, drag, and track cursor position.
- Keyboard Control: Input text, simulate key presses/combinations, and hold keys.
- Screen Operations: Capture the screen, detect active windows, and retrieve screen size.
- Clipboard Integration: Read, write, clear, and check clipboard content.
Link to the repository: Cheffromspace/MCPContro
9. Excel MCP Server
The Excel MCP Server provides powerful Excel file manipulation capabilities without requiring Microsoft Excel installation. Here are its key features:
- Workbook and Worksheet Management: Create, modify, and manage Excel workbooks and worksheets.
- Data Manipulation: Read, write, and transform data within Excel files.
- Formatting and Styling: Apply styles, formats, and customize the appearance of cells.
- Charts and Visualizations: Create charts and visual representations of data.
- Pivot Tables and Data Analysis: Generate pivot tables and perform advanced data analysis.
- Error Handling: Includes robust error handling with clear messages.
Link to the repository: haris-musa/excel-mcp-server
10. Fetch MCP Server
The Web Fetch MCP Server enables LLMs to retrieve and process content from web pages, converting HTML into markdown for easier consumption. Here are its key features:
- Content Retrieval: Fetch content from any URL on the internet.
- Markdown Conversion: Convert HTML content into markdown for simplified readability.
- Chunked Content Extraction: Use the start_index argument to extract content in chunks for large web pages.
- Raw Content Option: Retrieve raw HTML content without markdown conversion if needed.
- Customizable Output: Specify maximum content length for fetched data.
Link to the repository: servers/src/fetch
Why Use MCP Servers?
MCP servers are built to extend the capabilities of AI models by connecting them to external tools and data sources. They enable automation, boost efficiency, and seamlessly integrate with existing workflows. In the future, you will be able to ask AI to complete tasks for you—like preparing meals, checking the news, or even watering your plants. The AI will handle these tasks by sending structured information to other devices, ensuring smooth and coordinated operations.
Abid Ali Awan (@1abidaliawan) is a certified data scientist professional who loves building machine learning models. Currently, he is focusing on content creation and writing technical blogs on machine learning and data science technologies. Abid holds a Master’s degree in technology management and a bachelor’s degree in telecommunication engineering. His vision is to build an AI product using a graph neural network for students struggling with mental illness.