Discover Awesome MCP Servers

zei-world-mcp

Exposes ESG/CSR data from Zei World to LLMs, enabling sector browsing, company search, ESG score comparison, and detailed evaluation criteria exploration through natural language.

Global MCP Server

A modular MCP server that extends GitHub Copilot's capabilities through intelligent context compression and dynamic model routing for long-lived coding sessions.

Graft

Enables building agent-ready APIs that expose tools as both HTTP and MCP endpoints from a single server definition, with automatic OpenAPI, discovery docs, and interactive API reference.

smart-gym-mcp

Connects AI assistants to SmartGym on Mac, enabling users to review workouts, tweak routines, and build training programs in plain English, with changes syncing to iPhone and Apple Watch.

antigravity-browser-bridge

Enables MCP-compatible agents to securely control the user's already authenticated Chrome browser via explicit tab authorization and DOM-based actions.

Universal Agent Economy OS

Foundational MCP/A2A-native core platform for secure credential injection, native x402 micropayments, and Agent-to-Agent routing.

Weather and Stock Information MCP Server

An MCP server that provides real-time weather alerts, forecasts, stock market data, and news information through various API integrations.

MCP iOS Simulator Screenshot

Un servidor que captura capturas de pantalla del Simulador de iOS utilizando el Protocolo de Contexto del Modelo (MCP) y las guarda en un directorio especificado.

Wave MCP Server

Enables Wave invoicing operations including listing invoices, retrieving details, and generating branded PDFs directly from AI assistants.

Pulse Workflow MCP Server

Enables Claude Code to browse, create, edit, and publish Pulse workflows by exposing workflow operations as MCP tools.

Automated-Webflow

Okay, here's the translation of "Connecting MCP Servers to a Webflow project" into Spanish: **Conectar servidores MCP a un proyecto de Webflow** Or, a slightly more formal option: **Conexión de servidores MCP a un proyecto de Webflow**

UI Reviewer MCP

MCP-first UI/UX review layer for AI-generated frontends. Enables reviewing web pages via URL, capturing screenshots, extracting layout metrics, and generating structured repair plans for agents.

HippoDid MCP Server

Provides persistent, cloud-based memory for AI agents, letting them remember preferences, skills, and decisions across sessions via an MCP-compatible interface.

GitHub MCP Server

A comprehensive MCP server that exposes 73 GitHub API tools for managing repositories, pull requests, issues, actions, releases, search, and more, enabling natural language control of GitHub.

civitai-mcp-ultimate

A comprehensive MCP server for interacting with the Civitai API to search models, browse images with generation prompts, and manage downloads. It features extensive support for various base models like SDXL and Flux, includes NSFW filtering, and provides automated download commands for ComfyUI.

pilates_mcp

Local-first MCP server for Pilates centers that creates, searches, and reuses promotional poster HTML/CSS sources, records design feedback, and renders PNG/PDF artifacts with headless Chromium. It also includes a Schedule Workspace for managing group/private timetables and generating schedule posters.

web-ui-component-spec-mcp

Provides AI coding assistants with on-demand access to component specs, test scenarios, accessibility requirements, and build guides from the Web UI Component Specification.

MF Invoice MCP

An MCP server that integrates with MoneyForward Cloud Invoice API v3 to automate the creation and management of quotes and invoices. It supports Japanese invoice system compliance, allowing users to handle partners, items, and billing through natural language.

X API AI Agent – Personal Learning Project

Okay, I understand. You want me to be an AI agent capable of posting on X (formerly Twitter) using a Model Text Protocol (MCP) Server. Here's how I can function and what I need to operate effectively: **My Functionality:** 1. **Receive Instructions via MCP:** I will listen for instructions from an MCP server. These instructions will specify the content of the tweet, and potentially other parameters like scheduling, target audience, etc. 2. **Content Generation/Processing (If Needed):** * If the MCP server sends raw text, I'll simply prepare it for posting. * If the MCP server sends a prompt or request (e.g., "Write a tweet about the latest AI advancements"), I will use my language capabilities to generate a suitable tweet. I can also process and refine existing text to fit Twitter's character limits and style. 3. **Authentication and Authorization:** I will need to securely authenticate with the X API using credentials provided (and managed securely). This is crucial to ensure I have permission to post on the designated account. 4. **Posting to X:** I will use the X API to post the prepared tweet to the specified account. 5. **Error Handling and Reporting:** I will need to handle potential errors (e.g., API errors, authentication failures, rate limits) gracefully. I should log these errors and, ideally, report them back to the MCP server. 6. **Rate Limit Management:** I need to be aware of X's rate limits and avoid exceeding them. This might involve scheduling posts or pausing activity if limits are reached. 7. **Scheduling (Optional):** If the MCP server provides scheduling information, I can use a scheduling library (e.g., `schedule` in Python) to post tweets at specific times. **What I Need from You (or the System):** * **MCP Server Address and Protocol:** I need to know the address (IP or hostname) and port of the MCP server, as well as the specific protocol it uses for communication (e.g., HTTP, TCP sockets, etc.). I also need the message format expected by the MCP server. * **X API Credentials:** I need the necessary API keys, tokens, and secrets to authenticate with the X API. **These must be stored securely!** Consider using environment variables or a dedicated secrets management system. * **X Account Information:** The username or account ID of the X account I will be posting to. * **Error Handling Strategy:** How should I handle errors? Should I retry failed posts? Should I send error messages back to the MCP server? * **Rate Limit Handling Strategy:** How aggressively should I post? Should I prioritize certain posts over others? * **Programming Language and Libraries:** The programming language I will be implemented in (e.g., Python, Node.js, Java) and the libraries I can use (e.g., `tweepy` or `python-twitter` for the X API, a library for MCP communication, a scheduling library). * **Security Considerations:** How will you ensure that the MCP server is secure and that only authorized users can send instructions to me? How will you protect the X API credentials? **Example Implementation (Conceptual - Python):** ```python import tweepy import socket # For MCP communication (example) import json # For MCP message parsing (example) import os # For environment variables (secrets) import time # For rate limiting # --- Configuration (Load from environment variables or a config file) --- MCP_SERVER_HOST = os.environ.get("MCP_SERVER_HOST") MCP_SERVER_PORT = int(os.environ.get("MCP_SERVER_PORT", 12345)) X_API_KEY = os.environ.get("X_API_KEY") X_API_SECRET = os.environ.get("X_API_SECRET") X_ACCESS_TOKEN = os.environ.get("X_ACCESS_TOKEN") X_ACCESS_TOKEN_SECRET = os.environ.get("X_ACCESS_TOKEN_SECRET") # --- X API Authentication --- auth = tweepy.OAuthHandler(X_API_KEY, X_API_SECRET) auth.set_access_token(X_ACCESS_TOKEN, X_ACCESS_TOKEN_SECRET) api = tweepy.API(auth, wait_on_rate_limit=True) # Important for rate limiting def post_tweet(text): """Posts a tweet to X.""" try: api.update_status(text) print(f"Tweet posted successfully: {text}") except tweepy.TweepyException as e: print(f"Error posting tweet: {e}") # Handle the error (e.g., retry, log, report to MCP) def receive_mcp_message(): """Receives a message from the MCP server.""" try: with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: s.bind((MCP_SERVER_HOST, MCP_SERVER_PORT)) s.listen() conn, addr = s.accept() with conn: print(f"Connected by {addr}") data = conn.recv(1024) # Adjust buffer size as needed if not data: return None message = json.loads(data.decode()) # Assuming JSON format return message except Exception as e: print(f"Error receiving MCP message: {e}") return None def main(): """Main loop to receive instructions from MCP and post tweets.""" while True: message = receive_mcp_message() if message: if message.get("type") == "tweet": tweet_text = message.get("text") if tweet_text: post_tweet(tweet_text) else: print("Error: Tweet text is missing in the MCP message.") else: print("Unknown message type from MCP.") else: print("No message received from MCP. Sleeping...") time.sleep(10) # Check for messages every 10 seconds if __name__ == "__main__": main() ``` **Explanation of the Example:** * **Configuration:** Loads API keys and server addresses from environment variables (a good security practice). * **Authentication:** Uses `tweepy` to authenticate with the X API. `wait_on_rate_limit=True` is crucial to avoid exceeding rate limits. * **`post_tweet()`:** Posts the tweet using `api.update_status()`. Includes error handling. * **`receive_mcp_message()`:** Listens for connections on the specified port, receives data, and parses it as JSON (assuming the MCP server sends JSON messages). You'll need to adapt this to the specific MCP protocol. * **`main()`:** The main loop that continuously checks for messages from the MCP server and posts tweets accordingly. * **Error Handling:** Includes basic error handling for API errors and MCP communication errors. * **Rate Limiting:** `tweepy` helps with rate limiting, but you might need additional logic to avoid exceeding limits. **Important Considerations:** * **Security:** Protect your API keys! Never hardcode them directly into your code. Use environment variables or a secrets management system. * **Error Handling:** Implement robust error handling to deal with API errors, network issues, and other potential problems. * **Rate Limiting:** Be very careful about rate limits. X can suspend your account if you exceed them. * **MCP Protocol:** The most important part is understanding the exact protocol used by the MCP server. The example assumes JSON over TCP sockets, but you'll need to adapt it to your specific setup. * **Content Moderation:** Consider adding content moderation to prevent posting inappropriate or harmful content. * **Testing:** Thoroughly test your AI agent before deploying it to a production environment. This detailed explanation and example should give you a solid foundation for building your AI agent. Remember to adapt the code to your specific needs and environment. Good luck!

DWZ Short URL MCP Server

Enables AI assistants to create, manage, and analyze short URLs through complete URL shortening functionality. Supports batch operations, custom domains, click statistics, and comprehensive link management.

fireflies-mcp

MCP server and CLI for Fireflies.ai API to retrieve transcripts, search meetings, and generate summaries.

powerbi-mcp

MCP server for Power BI — local PBIX file analysis, cloud API operations, report export, and semantic model extraction from Power BI Desktop.

Concisr

Token compression for AI contexts, reducing token consumption by compressing conversation exchanges before they enter the LLM context window.

flow-team-mcp

Enables interaction with the Flow Team API through MCP clients, allowing endpoint discovery, API calls, and project management operations like listing projects and participants.

ESPN MCP Server

Provides comprehensive access to ESPN's sports APIs for live scores, team information, news, and statistics across NFL, NBA, MLB, NHL, college sports, and major soccer leagues with advanced filtering and real-time updates.

RA-Nested-MCP

A two-layer nested MCP server that acts as a telecom Revenue Assurance knowledge agent. It uses an orchestrator LLM (Llama 3.3 70B via Groq) to decompose questions, retrieve from a local vector store of RA documents, and synthesize answers, all running on free tiers.

Monday.com MCP Server

Enables AI assistants to interact with Monday.com workspaces, allowing retrieval of board lists, board details, item content, and user information through the Monday.com API.

MCP Server com SSE e Autenticação Bearer

Servidor MCP con SSE y autenticación Bearer.

Google Calendar MCP Server

Integrates Google Calendar with the Model Context Protocol to manage events, search agendas, and check availability. It features Home Assistant integration and SSE support for remote access through Claude Desktop.