Discover Awesome MCP Servers

GitHub MCP Bridge

A Model Context Protocol server that enables AI agents to securely access and interact with GitHub Enterprise data, providing access to enterprise users, organizations, emails, and license information.

docsonar

A local document search MCP server that indexes folders and provides hybrid keyword+semantic search, enabling users to chat with their documents via MCP clients like Claude Desktop or Claude Code.

codex3d

Enables natural language creation and refinement of Blender scenes through structured MCP tools, with persistent object identity, visual validation, and reversible edits.

AgentShare Price API

AgentShare delivers structured product search and pricing signals for AI agents over REST and MCP (Streamable HTTP). Responses include freshness & coverage metadata so agents can reason about data recency. API keys secure billed endpoints; public discovery at /agent.json and /mcp.json. Currently integrates connected marketplaces and affiliate feeds – roadmap expands to global e-commerce (AliExpre

suggest-skills

An MCP server that generates skill manifests from GitHub skills directories and provides tools to recommend and download AI agent skills.

gpt-image-mcp

An MCP server that enables Claude Code and Claude Desktop to generate and edit images using OpenAI's gpt-image-2 model, with results saved to disk.

email-send

Enables sending transactional emails with text or HTML body, delivery status, and message ID via Resend. Pay-per-call using x402 micropayments (USDC on Base L2), no API key or signup required.

protonmail-mcp

Enables AI clients to interact with ProtonMail accounts through the Proton Bridge using SMTP and IMAP protocols. Provides email management capabilities via secure local bridge connections.

LimaCharlie MCP

A local MCP server for the LimaCharlie security platform that provides investigation, administration, and content-review workflows via a broad read-only tool surface with explicit organization scoping and audit logging.

Paylocity MCP Server

Connects Claude Desktop to the Paylocity API to manage employee records, pay statements, and company headcount data through natural language. It includes automated data protection that redacts sensitive information like SSNs and bank account numbers before reaching the model.

Xiaohongshu (XHS) Creator Toolkit

An automation toolkit that enables AI-driven content publishing and creator data analysis for Xiaohongshu via the MCP protocol. It supports automated posting of image and video notes, cookie management, and performance metric tracking through natural language conversations.

acmt001-mcp

MCP server for ISO 20022 acmt.001 Account Opening (and companion acmt.* messages): message-type discovery, required-field lookup, JSON Schema introspection, IBAN/BIC/LEI validation, flat-record validation, and validated acmt XML generation.

Knowledge Graph Builder

Transforms text or web content into structured knowledge graphs using local AI models with MCP integration for persistent storage in Neo4j and Qdrant.

MCP Notion Server (@suncreation)

An MCP server that enables LLMs to interact with Notion workspaces via the Notion API, supporting page creation, database management, and content retrieval. It features markdown conversion to optimize token usage and enhanced error handling for more reliable workspace interactions.

RadMail MCP

An email operating system for AI agents that provides safe triage, drafting, and searching of emails with unbypassable hard-stops on money, banking changes, first-contact senders, and prompt injection.

Parakeet Transcription MCP Server

Transcribes audio/video files to text using NVIDIA's Parakeet TDT 0.6B V2 model, with optional timestamps; also retrieves model and system hardware info.

stock-analyzer-ai-mcp

Analyze stocks with financial ratios, comparisons, and sector performance data.

MCP Template

A template MCP server built with FastMCP framework that demonstrates basic tool implementation with a simple addition calculator example.

Google Calendar MCP Server

Servidor de Protocolo de Contexto del Modelo (MCP) que se integra con la API de Google Calendar.

textview-mcp

Connects AI assistants to TextView for persistent memory, enabling saving, searching, and retrieving notes across sessions.

Odoo 19 MCP Server

Provides tools to interact with Odoo 19's External JSON-2 API, enabling CRUD operations and complex queries on Odoo databases with multi-company support.

AI Agent with MCP

Okay, here's a basic outline and code snippets to help you create your first MCP (Model Context Protocol) server in a Playground environment. Keep in mind that MCP is a relatively new and evolving protocol, so the specific libraries and implementations might change. This example focuses on a simplified, conceptual approach. **Conceptual Overview** 1. **Choose a Language/Framework:** Python is a good choice for rapid prototyping and has libraries suitable for networking and data serialization. 2. **Define Your Model:** Decide what kind of model you want to serve. For a simple example, let's imagine a model that performs basic arithmetic (addition). 3. **Implement the MCP Server:** * Listen for incoming connections. * Receive MCP requests. * Parse the requests. * Execute the model (in our case, addition). * Format the response according to MCP. * Send the response. 4. **Implement a Simple MCP Client (for testing):** * Create a client to send requests to your server. * Receive and parse the responses. **Simplified Python Example (using `socket` and basic JSON)** ```python # server.py (This would run in your Playground) import socket import json HOST = '127.0.0.1' # Standard loopback interface address (localhost) PORT = 65432 # Port to listen on (non-privileged ports are > 1023) def handle_request(data): """ Simulates a simple model that performs addition. Assumes the data is a JSON string with 'a' and 'b' keys. """ try: request = json.loads(data.decode('utf-8')) a = request.get('a') b = request.get('b') if a is None or b is None: return json.dumps({"error": "Missing 'a' or 'b' parameter"}).encode('utf-8') try: result = a + b response = {"result": result} return json.dumps(response).encode('utf-8') except TypeError: return json.dumps({"error": "Invalid 'a' or 'b' value (must be numbers)"}).encode('utf-8') except json.JSONDecodeError: return json.dumps({"error": "Invalid JSON"}).encode('utf-8') with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((HOST, PORT)) s.listen() print(f"Server listening on {HOST}:{PORT}") conn, addr = s.accept() with conn: print(f"Connected by {addr}") while True: data = conn.recv(1024) # Receive up to 1024 bytes if not data: break # Client disconnected response = handle_request(data) conn.sendall(response) # Send the response back to the client ``` ```python # client.py (This would run in a separate Playground or terminal) import socket import json HOST = '127.0.0.1' # The server's hostname or IP address PORT = 65432 # The port used by the server with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.connect((HOST, PORT)) # Example request (MCP-like, but simplified) request_data = {"a": 5, "b": 3} request_json = json.dumps(request_data).encode('utf-8') s.sendall(request_json) data = s.recv(1024) print('Received:', repr(data.decode('utf-8'))) ``` **Explanation and Key Points:** * **`server.py`:** * Sets up a basic TCP socket server. * `handle_request()`: This is where your "model" logic goes. In this example, it's a simple addition function. It receives JSON data, parses it, performs the addition, and returns a JSON response. Error handling is included. * The server listens for connections, accepts a connection, and then enters a loop to receive data, process it, and send a response. * **`client.py`:** * Creates a TCP socket client. * Connects to the server. * Constructs a JSON request (representing an MCP-like request). * Sends the request to the server. * Receives the response and prints it. * **JSON for Serialization:** JSON is used for encoding and decoding the requests and responses. This is a common and relatively simple way to handle data serialization. * **Error Handling:** Basic error handling is included in the `handle_request` function to catch invalid JSON, missing parameters, and type errors. * **Simplified MCP:** This is *not* a full MCP implementation. It's a simplified example to illustrate the basic concepts. A real MCP implementation would involve more complex message structures, metadata, and potentially other protocols for data transfer. * **Playground Considerations:** Make sure your Playground environment allows network connections. Some online Playgrounds might have restrictions. If you have issues, try running the server and client on your local machine. * **Running the Code:** 1. Run `server.py` in one Playground or terminal window. 2. Run `client.py` in another Playground or terminal window. 3. The client should connect to the server, send the request, and print the response. **To make this more like a real MCP server, you would need to:** * **Define a formal MCP message structure:** MCP has specific requirements for the format of requests and responses, including metadata and data encoding. You'd need to adhere to those specifications. * **Implement a more sophisticated model:** Replace the simple addition with a more complex machine learning model. You might use libraries like TensorFlow, PyTorch, or scikit-learn. * **Handle different data types:** MCP needs to support various data types (images, text, etc.). You'd need to implement appropriate serialization and deserialization methods. * **Add authentication and authorization:** Secure your server to prevent unauthorized access. * **Consider performance:** Optimize your code for speed and efficiency, especially if you're serving a high volume of requests. **Important Considerations for Playgrounds:** * **Network Access:** Many online Playgrounds have limited or no network access. If you can't get the server and client to connect, it's likely a network restriction. Try running the code locally on your machine. * **Dependencies:** Make sure your Playground environment has the necessary libraries installed (e.g., `json`). If not, you might need to install them using `pip` or a similar package manager. * **File System Access:** Some Playgrounds might restrict file system access. If you need to load model files, you might need to find alternative ways to store and access them (e.g., using cloud storage). This example provides a starting point. You'll need to research the specific MCP specifications and adapt the code to your particular model and requirements. Remember to consult the official MCP documentation and any relevant libraries for more detailed information.

hybrid-rag-mcp

An MCP server that provides a search_docs tool with hybrid retrieval (BM25 + dense vectors) and cross-encoder reranking, backed by evaluation, prompt-injection guardrails, and OpenTelemetry tracing.

mcp-funcheap

Enables querying San Francisco Bay Area events from Funcheap, including event details, tags, and categories.

macos-mcp

Local MCP server for macOS native apps: Mail, Calendar, Reminders, Notes, Messages, and Contacts. Enables reading and organizing your Mac life through a single stdio process using AppleScript/JXA.

neurodivergent-memory

MCP server for knowledge graphs designed around neurodivergent thinking patterns, organizing memories into five districts with BM25 ranking and bidirectional connections.

laptop-care

A laptop maintenance agent for Claude Desktop that inspects your machine, explains findings, recommends actions, and remembers history, with safety gates enforced by the server.

@langapi/mcp-server

MCP server for AI-powered translation management in i18n projects, enabling automated locale detection, translation status checks, and sync via LangAPI.

Gluestack Components MCP Server

Enables integration of Gluestack components with Claude Desktop and Cursor IDE for UI development.

Basic MCP

A simple MCP server built with FastMCP for experimentation and learning purposes. Includes basic web tools like article fetching and serves as a human-readable template for building custom MCP servers.