Discover Awesome MCP Servers

MCP App Test Value Picker Server

A debug tool for platform developers to verify that their MCP App hosts correctly implement protocol specifications like context injection and message sending. It provides a blind selection test to ensure seamless end-to-end communication between the UI, host, and model.

Multi-Agent RAG MCP Server

A legal-tech focused system that coordinates specialized agents for document classification, deadline extraction from Spanish legal texts, and strategic business intelligence. It integrates with Claude via MCP to provide semantic document search and automated deadline tracking using a Supabase vector database.

Quant Companion MCP

Provides real-time options analytics, pricing with Greeks, Monte Carlo simulations, volatility analysis, strategy backtesting, and risk metrics using actual market data from Yahoo Finance and Polygon.io.

nonmem-mcp-server

An MCP server for NONMEM pharmacometric modeling that provides structured access to model parsing, execution, and results analysis. It enables users to perform diagnostics, manage PsN workflows, and translate models to mrgsolve for PK simulations through natural language.

KPC Component Library MCP Server

AI code generation service for KPC component library that solves hallucination problems by providing accurate component APIs, validation rules, and usage examples.

Logseq MCP Server

Enables AI assistants like Claude to directly read, write, search, and navigate your local Logseq knowledge graph, including managing journals, pages, backlinks, and page relationships without manual copy-pasting.

mcp-servers

Repository for Model Context Protocol Servers

Draw Things MCP Server

Enables LLMs to generate and transform images locally on Mac using Stable Diffusion through the Draw Things app, supporting text-to-image and image-to-image generation with Apple Silicon acceleration.

Investment Memorandum Processor MCP Server

An MCP server that automates extraction of structured financial data from investment memorandums and generates standardized PowerPoint presentations through a RESTful API.

Job URL Analyzer MCP Server

A FastAPI-based microservice that analyzes job URLs and extracts detailed company information by crawling job postings and company websites, with data enrichment from external providers.

MCP Domotica Backend

Enables control and management of smart home devices across multiple rooms through specialized MCP servers. Supports lights, thermostats, fans, and ovens with room-specific rules and automatic persistence.

Docker MCP Server

Enables natural language interaction with Docker commands and operations. Supports container management, image operations, system information, and Docker Compose through conversational requests.

Ingrids Reisetjenester

Enables intelligent travel planning by combining weather forecasts and route calculations for destinations. Provides personalized travel recommendations based on weather conditions and supports trip planning with persistent conversation memory.

Task Manager MCP

Enables intelligent task management with status tracking, dependency resolution, and automatic next task discovery based on preconditions and priorities. Supports hierarchical task structures with subtasks and flexible JSON-based configuration.

jensenify-mcp

An MCP server that injects millions of tokens from classical literature into AI engineering interactions to provide deep humanistic context for technical decisions. It allows users to consult works like the Iliad or Shakespeare for engineering advice while helping them maximize their compute spending.

FoundryVTT MCP Server

A Model Context Protocol server that integrates with FoundryVTT, allowing AI assistants to interact with tabletop gaming sessions through natural language to query actors, roll dice, generate content, and manage game worlds.

Search Stock News MCP Server

Provides real-time stock news search capabilities via Tavily API, allowing MCP clients to retrieve filtered and customized stock news with various search parameters.

vivioo-mcp

Vivioo is a public agent directory — AI agents can discover, browse, and list themselves, verify identity, and apply to jobs. 8 tools, no authentication required.

Schedulia MCP

A meeting scheduling assistant that enables users to view schedules, manage incoming requests, and send meeting invitations via the Schedulia API. It facilitates seamless coordination of meeting times and participant management through natural language commands.

MediaWiki MCP Server

A tool that enables AI assistants like Claude to interact with MediaWiki instances by retrieving page content, performing searches, and analyzing wiki information through the MediaWiki API.

Ollama Pydantic Project

承知いたしました。以下に、ローカルのOllamaモデルとMCPサーバー統合を使用したPydanticエージェントのサンプルプロジェクトを作成します。 **プロジェクト構成:** ``` pydantic_ollama_mcp_agent/ ├── agent.py # Pydanticエージェントの定義 ├── ollama_client.py # Ollamaモデルとのインターフェース ├── mcp_client.py # MCPサーバーとのインターフェース ├── main.py # メインの実行ファイル └── requirements.txt # 依存関係 ``` **1. `requirements.txt`:** ``` pydantic ollama requests # 必要に応じて、MCPサーバーのクライアントライブラリを追加 ``` **2. `agent.py`:** ```python from pydantic import BaseModel, Field from typing import List class Task(BaseModel): description: str = Field(..., description="タスクの説明") priority: int = Field(1, description="タスクの優先度 (1: 高, 2: 中, 3: 低)") class AgentResponse(BaseModel): action: str = Field(..., description="実行するアクション") reasoning: str = Field(..., description="アクションの理由") arguments: dict = Field(..., description="アクションに必要な引数") class PydanticAgent: def __init__(self, ollama_client, mcp_client): self.ollama_client = ollama_client self.mcp_client = mcp_client def run(self, task: Task) -> AgentResponse: """ タスクを受け取り、Ollamaモデルを使ってアクションを決定し、MCPサーバーに実行を依頼します。 """ prompt = f"タスク: {task.description}\n優先度: {task.priority}\n実行すべきアクションを決定してください。JSON形式でAgentResponseを返してください。" response_text = self.ollama_client.generate_text(prompt) try: agent_response = AgentResponse.parse_raw(response_text) print(f"Ollamaからのレスポンス: {agent_response}") # デバッグ用 # MCPサーバーにアクションを依頼 mcp_response = self.mcp_client.execute_action(agent_response.action, agent_response.arguments) print(f"MCPサーバーからのレスポンス: {mcp_response}") # デバッグ用 return agent_response except Exception as e: print(f"エラー: {e}") return AgentResponse(action="error", reasoning=f"エラーが発生しました: {e}", arguments={}) ``` **3. `ollama_client.py`:** ```python import ollama class OllamaClient: def __init__(self, model_name="llama2"): # モデル名を指定 self.model_name = model_name def generate_text(self, prompt: str) -> str: """ Ollamaモデルにプロンプトを送信し、テキストを生成します。 """ try: response = ollama.generate(model=self.model_name, prompt=prompt) return response['response'] except Exception as e: print(f"Ollamaエラー: {e}") return "{'action': 'error', 'reasoning': 'Ollamaモデルとの通信に失敗しました', 'arguments': {}}" ``` **4. `mcp_client.py`:** ```python import requests import json class MCPClient: def __init__(self, mcp_server_url="http://localhost:8000"): # MCPサーバーのURLを指定 self.mcp_server_url = mcp_server_url def execute_action(self, action: str, arguments: dict) -> dict: """ MCPサーバーにアクションの実行を依頼します。 """ url = f"{self.mcp_server_url}/execute" payload = {"action": action, "arguments": arguments} headers = {'Content-type': 'application/json'} try: response = requests.post(url, data=json.dumps(payload), headers=headers) response.raise_for_status() # HTTPエラーをチェック return response.json() except requests.exceptions.RequestException as e: print(f"MCPサーバーエラー: {e}") return {"status": "error", "message": f"MCPサーバーとの通信に失敗しました: {e}"} ``` **5. `main.py`:** ```python from agent import PydanticAgent, Task from ollama_client import OllamaClient from mcp_client import MCPClient def main(): # クライアントの初期化 ollama_client = OllamaClient() mcp_client = MCPClient() # エージェントの初期化 agent = PydanticAgent(ollama_client, mcp_client) # タスクの定義 task = Task(description="今日の天気を調べて、雨が降る場合は傘を持っていくようにリマインダーを設定してください。", priority=1) # エージェントの実行 response = agent.run(task) print(f"最終結果: {response}") if __name__ == "__main__": main() ``` **使い方:** 1. **Ollamaのインストールとモデルのダウンロード:** Ollamaをインストールし、`llama2`モデルをダウンロードします。 `ollama pull llama2` 2. **MCPサーバーの準備:** MCPサーバーを起動し、`http://localhost:8000`でアクセスできるようにします。 MCPサーバーは、`execute`エンドポイントを持ち、アクション名と引数を受け取り、実行結果をJSONで返す必要があります。 (MCPサーバーの実装は、この例では提供されていません。 FlaskやFastAPIなどで実装する必要があります。) 3. **依存関係のインストール:** `pip install -r requirements.txt` 4. **実行:** `python main.py` **重要な点:** * **Ollamaモデルの選択:** `OllamaClient`の`model_name`を、使用するOllamaモデルの名前に変更してください。 * **MCPサーバーのURL:** `MCPClient`の`mcp_server_url`を、MCPサーバーのURLに変更してください。 * **MCPサーバーの実装:** この例では、MCPサーバーの実装は提供されていません。 MCPサーバーは、エージェントから送られてくるアクション名と引数を受け取り、実際にアクションを実行し、結果をJSONで返す必要があります。 * **エラーハンドリング:** エラーハンドリングは最小限に留めています。 より堅牢なアプリケーションでは、より詳細なエラーハンドリングが必要です。 * **プロンプトエンジニアリング:** Ollamaモデルに渡すプロンプトは、モデルの性能に大きく影響します。 より良い結果を得るためには、プロンプトを調整する必要があります。 * **セキュリティ:** 本番環境で使用する場合は、セキュリティを考慮する必要があります。 **MCPサーバーの例 (Flask):** ```python from flask import Flask, request, jsonify app = Flask(__name__) @app.route('/execute', methods=['POST']) def execute(): data = request.get_json() action = data['action'] arguments = data['arguments'] print(f"アクション: {action}, 引数: {arguments}") # ここで実際のアクションを実行するロジックを実装 if action == "set_reminder": # リマインダーを設定する処理 message = f"リマインダーを設定しました: {arguments}" status = "success" elif action == "get_weather": # 天気を取得する処理 message = "今日の天気は晴れです" status = "success" else: message = "不明なアクションです" status = "error" return jsonify({"status": status, "message": message}) if __name__ == '__main__': app.run(debug=True, port=8000) ``` このサンプルプロジェクトは、Pydanticエージェント、Ollamaモデル、MCPサーバーを統合するための基本的なフレームワークを提供します。 必要に応じて、機能を拡張し、特定の要件に合わせてカスタマイズしてください。

schwab-mcp

An MCP server that connects to the Charles Schwab brokerage API to enable account management, market data retrieval, and secure order execution. It includes a Claude Code plugin with specialized skills for optimized tool selection and safety-focused confirmation patterns for trading.

Oura MCP

Provides LLMs with access to Oura Ring health data including sleep metrics, activity tracking, heart rate, readiness scores, and other wellness insights through the Oura API v2.

Agent HTTP

Provides an HTTP API for Claude Code using a native MCP channel to enable stable message exchange without terminal screen-scraping. It allows users to programmatically send messages, retrieve conversation history, and stream real-time events via standard web protocols.

ZugaShield

A 7-layer security system for AI agents that detects and blocks prompt injection, data exfiltration, and malicious tool calls. It enables real-time scanning of inputs, outputs, and tool definitions to protect agentic workflows from emerging AI-specific threats.

Google Business Profile Review MCP Server

Enables AI assistants to fetch, analyze, and respond to Google Business Profile reviews with AI-generated replies through secure OAuth integration with Google's My Business API.

issue-tracker-mcp

A minimal MCP server that enables AI assistants to manage a kanban issue board. It provides tools for listing, creating, updating, and deleting issues with support for both local development and team deployments.

Bootpay Developer Docs MCP Server

Enables AI coding tools to search and retrieve Bootpay payment and commerce developer documentation, including integration guides and customer service manuals. It facilitates tasks such as payment linking, billing key issuance, and webhook configuration through natural language queries.