Discover Awesome MCP Servers

MCP Weather Server

Enables AI agents to access real-time and historical weather data through multiple weather APIs including OpenMeteo, Tomorrow.io, and OpenWeatherMap. Provides comprehensive meteorological information including current conditions, forecasts, historical data, and weather alerts.

claude-session-continuity-mcp

Zero-config session continuity for Claude Code. Automatically captures and restores project context across sessions using Claude Hooks.

MORAGENT AI Agent Studio

Turns Claude Code into an AI Agent Studio with a guided menu and 11 tools to design, create, and manage multi-agent projects without coding.

Torify — Japanese Locale APIs for AI Agents

Torify gives AI agents the Japanese locale primitives that standard libraries lack — imperial era date conversion (wareki), qualified invoice number validation with NTA registry lookup, corporate number lookup (法人番号), postal code resolution, name romanization (Hepburn), and kanji-to-kana conversion via Yahoo! JLP. 31 endpoints total. No authentication required for MCP. Pay-per-call $0.02/call via

Felix MCP (Smithery)

A lightweight MCP server with greeting, random number, weather, and OpenAI-powered text summarization tools, deployable on Smithery.

alaya

Enables Claude Code to serve as the primary interface for a personal knowledge vault (zk or Obsidian), allowing full read, write, search, and synthesis operations on notes.

MCP Manager

A gateway that manages multiple MCP servers behind one endpoint, providing namespaced tool routing and supervision. Currently bundles an Arxiv MCP server for paper search and retrieval.

Pokemon Champions Battle Assistant

Enables live battle statistics queries and precise damage calculations for Pokemon Champions, supporting Japanese and English names, multiple formats, and environmental factors.

tasksync-mcp

MCP server to give new instructions to agent while its working. It uses the get_feedback tool to collect your input from the feedback.md file in the workspace, which is sent back to the agent when you save.

ebay-mcp

An MCP server that lets an LLM search and inspect eBay listings through the official Browse API.

@theyahia/voximplant-mcp

MCP server for Voximplant API enabling calls, SMS, recordings, scenarios, and rules management with 11 tools and 2 skills.

MarkdownMCP

Live markdown preview for LLM agents. Call a single tool to watch a markdown file and it opens a browser UI that streams rendered updates, changes, and history.

Task API MCP

Enables AI clients to create and manage tasks via a local REST API by converting natural language into HTTP requests through the Model Context Protocol.

Stamp it

An MCP server that adds full-screen text or image watermarks to images with intelligent color adaptation and multi-language support.

PostgreSQL MCP Server

Enables LLMs to interact deeply with PostgreSQL databases—query data, manage schema, analyze performance, and administer the database.

Enterprise Template Generator

Enables generation of enterprise-grade software templates with built-in GDPR/Swedish compliance validation, workflow automation for platform migrations, and comprehensive template management through domain-driven design principles.

Bocha Search MCP

Un motor de búsqueda centrado en la inteligencia artificial que permite a las aplicaciones de IA acceder a conocimiento de alta calidad proveniente de miles de millones de páginas web y fuentes de contenido del ecosistema en diversos dominios, incluyendo el clima, noticias, enciclopedias, información médica, billetes de tren e imágenes.

Build

Okay, I can help you understand how to create different MCP (Model Configuration Protocol) servers using the TypeScript SDK. However, I need a little more information to give you the *most* helpful and specific answer. Please tell me: 1. **Which MCP SDK are you using?** There are several possibilities. For example: * Are you referring to a specific cloud provider's MCP service (e.g., Google Cloud's Model Registry, AWS SageMaker Model Registry, Azure Machine Learning Model Registry)? If so, please specify which one. * Are you using a more general-purpose MCP library or framework? If so, please provide the name of the library/framework. * Are you building your own MCP server from scratch? 2. **What do you mean by "different"?** What aspects of the MCP server do you want to customize or differentiate? For example: * **Different data storage:** Do you want to store model configurations in different databases (e.g., PostgreSQL, MongoDB, a file system)? * **Different authentication/authorization:** Do you need different ways to authenticate users or control access to model configurations? * **Different APIs:** Do you want to expose different endpoints or use a different API style (e.g., REST, gRPC)? * **Different model configuration formats:** Do you want to support different formats for defining model configurations (e.g., JSON, YAML, Protobuf)? * **Different model deployment targets:** Do you want to manage models deployed to different environments (e.g., cloud, edge, on-premise)? * **Different metadata:** Do you want to store different metadata about your models? 3. **What is your current level of understanding?** Are you just starting out, or have you already tried something? Providing code snippets of what you've tried will help me understand your current progress and provide more targeted guidance. **General Concepts and Approaches (Without Specific SDK Information)** Assuming you're building something from scratch or using a more general-purpose library, here's a breakdown of the common elements involved in creating an MCP server and how you might differentiate them: * **Data Model:** The core of an MCP server is the data model that represents model configurations. This typically includes: * Model name/ID * Model version * Model metadata (e.g., description, author, creation date) * Model parameters (e.g., learning rate, batch size) * Model artifacts (e.g., the trained model file, code dependencies) * Deployment information (e.g., target environment, resource requirements) You can differentiate MCP servers by using different data models. For example, one server might focus on deep learning models and store specific information about neural network architectures, while another might focus on traditional machine learning models and store information about feature engineering pipelines. * **Storage Layer:** The storage layer is responsible for persisting model configurations. Common options include: * **Relational databases (e.g., PostgreSQL, MySQL):** Good for structured data and complex queries. * **NoSQL databases (e.g., MongoDB, Cassandra):** Good for flexible schemas and scalability. * **Object storage (e.g., AWS S3, Google Cloud Storage):** Good for storing large model artifacts. * **File system:** Simple but less scalable. You can differentiate MCP servers by using different storage layers. For example, one server might use PostgreSQL for metadata and S3 for model artifacts, while another might use MongoDB for everything. * **API Layer:** The API layer provides an interface for clients to interact with the MCP server. Common options include: * **REST:** A widely used API style based on HTTP. * **gRPC:** A high-performance API style based on Protocol Buffers. * **GraphQL:** A query language for APIs. You can differentiate MCP servers by using different API layers. For example, one server might expose a REST API for easy integration with web applications, while another might expose a gRPC API for high-performance communication between microservices. * **Authentication and Authorization:** These mechanisms control access to the MCP server. Common options include: * **API keys:** Simple but less secure. * **OAuth 2.0:** A widely used standard for delegated authorization. * **Role-based access control (RBAC):** Assigning permissions to roles and then assigning roles to users. You can differentiate MCP servers by using different authentication and authorization mechanisms. For example, one server might use API keys for internal access and OAuth 2.0 for external access. * **Deployment:** How the MCP server is deployed and managed. * **Cloud-based:** Deployed on cloud platforms like AWS, Azure, or GCP. * **On-premise:** Deployed on your own infrastructure. * **Containerized (Docker, Kubernetes):** Provides portability and scalability. You can differentiate MCP servers by deploying them in different environments. **Example (Conceptual - No Specific SDK)** Let's say you want to create two MCP servers: * **MCP Server 1:** For managing TensorFlow models deployed to Google Cloud. It uses PostgreSQL for metadata and Google Cloud Storage for model artifacts. It exposes a REST API and uses Google Cloud IAM for authentication. * **MCP Server 2:** For managing PyTorch models deployed to edge devices. It uses MongoDB for metadata and a local file system for model artifacts. It exposes a gRPC API and uses API keys for authentication. In this case, you would need to: 1. Define different data models for TensorFlow and PyTorch models. 2. Implement different storage layers using PostgreSQL/GCS and MongoDB/file system. 3. Implement different API layers using REST and gRPC. 4. Implement different authentication mechanisms using Google Cloud IAM and API keys. **TypeScript Code Snippet (Illustrative - Requires Specific SDK)** ```typescript // This is a very high-level example and needs to be adapted to your specific SDK. // Example of defining a data model (simplified) interface TensorFlowModelConfig { name: string; version: string; architecture: string; learningRate: number; gcsArtifactPath: string; // Path to the model in Google Cloud Storage } interface PyTorchModelConfig { name: string; version: string; modelDefinition: string; // Path to the model definition file batchSize: number; localArtifactPath: string; // Path to the model on the local file system } // Example of a simplified API endpoint (using Express.js) import express from 'express'; const app = express(); const port = 3000; app.get('/tensorflow/models/:name', (req, res) => { // Logic to retrieve TensorFlow model config from PostgreSQL and GCS const modelName = req.params.name; // ... (Retrieve from database and storage) const modelConfig: TensorFlowModelConfig = { name: modelName, version: "1.0", architecture: "CNN", learningRate: 0.001, gcsArtifactPath: "gs://my-bucket/my-model.pb" }; res.json(modelConfig); }); app.listen(port, () => { console.log(`Example app listening at http://localhost:${port}`); }); ``` **Next Steps** Please provide the information I requested at the beginning of this response (specifically, the MCP SDK you're using and what you mean by "different"). With that information, I can give you much more specific and helpful guidance, including code examples tailored to your situation.

mcp-dostuff

Enables listing metros and fetching events from the DoStuff network via MCP.

Iris MCP Server

A multi-backend gateway that enables access to various services like Google Drive and Notion through a single MCP connector. It currently provides comprehensive Google Drive integration for reading, writing, and managing files and folders.

Prompt Bookmarks

Enables users to organize, search, and manage a shared library of prompts across AI tools via the Model Context Protocol. It supports hierarchical folder organization, tagging, and template variable substitution for dynamic prompt generation.

Android Puppeteer

Enables AI agents to interact with Android devices through visual UI element detection and automated interactions. Provides comprehensive Android automation capabilities including touch gestures, text input, screenshots, and video recording via uiautomator2.

hive-exp

An MCP server enabling AI agents to record, query, and share structured problem-solving experiences with human review and confidence decay.

mcp-units

MCP server for converting cooking measurements (volume, weight, temperature) between common units like ml, cup, g, oz, and Celsius/Fahrenheit.

MCP-Discord

Enables AI assistants to interact with Discord servers through a bot, supporting channel management, messaging, forum operations, reactions, and webhooks.

MCP Document Indexer

Enables real-time indexing and semantic search of local documents (PDF, Word, text, Markdown, RTF) using vector embeddings and local LLMs. Monitors folders for changes and provides natural language search capabilities through Claude Desktop integration.

mcp-homebrew-formulae

Query Homebrew formulae, casks, and install analytics through natural language or direct tool calls.

Apple Maps MCP Server

Search Apple Maps for businesses with Apple ratings and aggregated Yelp and TripAdvisor reviews. Useful for lead generation, restaurant research, and competitive analysis.

Test Generator MCP Server

Enables automatic generation of test scenarios from user stories uploaded to Claude desktop. Leverages MCP integration to streamline the test case creation process for development workflows.