Discover Awesome MCP Servers

filesystem-mcp

Um servidor MCP baseado em TypeScript que implementa um sistema de notas simples, permitindo que usuários criem, acessem e gerem resumos de notas de texto via URIs e ferramentas.

svelte-llm

Svelte and SvelteKit developer odcumentation

Elysia MCP Starter

A template for building Model Context Protocol servers using Elysia and Bun runtime, enabling LLM clients like Claude Desktop and Cody to access custom tools, prompts, and data resources.

MCP Todo.txt Integration

A server implementation that enables LLMs to programmatically manage tasks in Todo.txt files using the Model Context Protocol (MCP), supporting operations like adding, completing, deleting, listing, searching, and filtering tasks.

JSer.info MCP Server

A Model Context Protocol server that provides search and retrieval capabilities for JSer.info's JavaScript resource database, enabling access to items, posts, product information, and timeline data through various specialized tools.

Plane MCP Server

A Model Context Protocol (MCP) server that enables LLMs to interact with Plane.so, allowing them to manage projects and issues through Plane's API. Using this server, LLMs like Claude can directly interact with your project management workflows while maintaining user control and security.

MCP

Okay, let's break down how you can configure an MCP (presumably, you mean a **Monitoring and Control Platform** or a similar system) to view company information and stock prices using Claude (likely referring to the **Anthropic Claude AI model**). This involves several steps and considerations. I'll outline a general approach, and you'll need to adapt it to your specific MCP and its capabilities. **High-Level Overview** The core idea is to: 1. **Gather Data:** Get company information and stock prices from reliable sources (APIs, databases, etc.). 2. **Send Data to Claude:** Format the data and send it to the Claude API with a clear prompt. 3. **Receive and Parse Claude's Response:** Claude will analyze the data and provide insights. You need to parse this response. 4. **Display in MCP:** Integrate the parsed information into your MCP's interface. **Detailed Steps** **1. Data Acquisition** * **Company Information:** * **APIs:** Use APIs like the Crunchbase API, Clearbit API, or similar services that provide company profiles, funding information, employee counts, industry, etc. These usually require an API key and have usage limits. * **Databases:** If your company has its own database of company information, you can directly query that. * **Web Scraping (Use with Caution):** As a last resort, you *could* scrape websites like LinkedIn or company websites. However, web scraping is fragile (websites change), and you need to respect robots.txt and terms of service. It's generally better to use APIs. * **Stock Prices:** * **Financial APIs:** Use APIs like Alpha Vantage, IEX Cloud, Finnhub, or Yahoo Finance API (though Yahoo Finance's API is less reliable than it used to be). These provide real-time or near real-time stock prices, historical data, and other financial metrics. Again, you'll need an API key. * **Data Providers:** Consider professional data providers like Refinitiv or Bloomberg if you need very high-quality, low-latency data, but these are significantly more expensive. **2. Data Preparation and Formatting** * **Data Cleaning:** Clean the data you retrieve. Handle missing values, inconsistencies, and errors. * **Data Aggregation:** Combine the company information and stock price data into a single data structure (e.g., a Python dictionary or a JSON object) for each company you want to analyze. * **Prompt Engineering:** This is crucial for getting useful results from Claude. Craft a clear and specific prompt that tells Claude what you want it to do. Here's an example: ```python prompt = f""" Analyze the following information about {company_name}: Company Description: {company_description} Industry: {company_industry} Employee Count: {company_employee_count} Current Stock Price: {stock_price} Previous Day's Closing Price: {previous_close} Provide a brief summary of the company, its current financial situation based on the stock price, and potential risks or opportunities. Focus on key insights that would be relevant to a business analyst. Keep the response concise (under 200 words). """ ``` * **Variables:** Replace the placeholders (e.g., `{company_name}`, `{stock_price}`) with the actual data you've collected. * **Instructions:** Clearly tell Claude what you want it to do (summarize, analyze, identify risks, etc.). * **Context:** Provide enough context for Claude to understand the data. * **Constraints:** Set limits on the response length to avoid overly verbose outputs. * **Role:** You can even specify a role for Claude to play, such as "Act as a financial analyst..." **3. Sending Data to Claude (Using the API)** * **Install the Anthropic Python Library:** ```bash pip install anthropic ``` * **API Key:** You'll need an API key from Anthropic. Get one from their website after signing up for an account. * **Code Example (Python):** ```python import anthropic import os # Replace with your actual API key ANTHROPIC_API_KEY = os.environ.get("ANTHROPIC_API_KEY") # Best practice: store API key in an environment variable client = anthropic.Anthropic(api_key=ANTHROPIC_API_KEY) def get_claude_response(prompt): try: completion = client.completions.create( model="claude-v1.3", # Or a newer Claude model prompt=f"{anthropic.HUMAN_PROMPT} {prompt}{anthropic.AI_PROMPT}", max_tokens_to_sample=500, # Adjust as needed ) return completion.completion except Exception as e: print(f"Error calling Claude API: {e}") return None # Example usage: company_name = "Apple Inc." company_description = "Designs, develops, and sells consumer electronics, computer software, and online services." company_industry = "Technology" company_employee_count = 164000 stock_price = 175.00 previous_close = 173.50 prompt = f""" Analyze the following information about {company_name}: Company Description: {company_description} Industry: {company_industry} Employee Count: {company_employee_count} Current Stock Price: {stock_price} Previous Day's Closing Price: {previous_close} Provide a brief summary of the company, its current financial situation based on the stock price, and potential risks or opportunities. Focus on key insights that would be relevant to a business analyst. Keep the response concise (under 200 words). """ claude_response = get_claude_response(prompt) if claude_response: print(f"Claude's Analysis for {company_name}:\n{claude_response}") else: print("Failed to get a response from Claude.") ``` * **`anthropic.HUMAN_PROMPT` and `anthropic.AI_PROMPT`:** These are special tokens that help Claude understand the structure of the conversation. Always include them. * **`model`:** Specify the Claude model you want to use (e.g., "claude-v1.3", "claude-v2"). Check the Anthropic documentation for the latest models. * **`max_tokens_to_sample`:** Controls the maximum length of Claude's response. Adjust this based on your needs. * **Error Handling:** Include `try...except` blocks to handle potential errors when calling the API. **4. Parsing Claude's Response** * **Text Extraction:** The `completion.completion` attribute contains the text response from Claude. * **Structured Data (Optional):** If you want Claude to return structured data (e.g., JSON), you can modify your prompt to request it. However, Claude is not always perfect at generating valid JSON, so you might need to use regular expressions or other parsing techniques to extract the data reliably. For example: ```python prompt = f""" Analyze the following information about {company_name}: Company Description: {company_description} Industry: {company_industry} Employee Count: {company_employee_count} Current Stock Price: {stock_price} Previous Day's Closing Price: {previous_close} Provide a JSON object with the following keys: - summary: A brief summary of the company. - financial_outlook: A brief assessment of the company's financial situation. - risks: Potential risks. - opportunities: Potential opportunities. Example JSON: {{ "summary": "...", "financial_outlook": "...", "risks": "...", "opportunities": "..." }} """ ``` Then, you would try to parse the `claude_response` as JSON: ```python import json try: data = json.loads(claude_response) print(f"Summary: {data['summary']}") print(f"Financial Outlook: {data['financial_outlook']}") except json.JSONDecodeError: print("Could not parse Claude's response as JSON.") # Handle the error (e.g., use a regular expression to extract the data) ``` **5. Integration with Your MCP** * **MCP API/SDK:** Your MCP likely has an API or SDK that allows you to integrate external data sources. Consult your MCP's documentation for details. * **Data Visualization:** Use your MCP's charting and visualization tools to display the company information, stock prices, and Claude's analysis. * **Alerting:** Configure alerts based on stock price changes or specific insights from Claude (e.g., "Alert if Claude identifies a significant risk for Apple"). **Example Architecture** ``` [Data Sources (APIs, Databases)] --> [Data Aggregation & Formatting (Python Script)] --> [Claude API] --> [Response Parsing (Python Script)] --> [MCP API] --> [MCP Dashboard] ``` **Important Considerations** * **API Costs:** Be aware of the costs associated with using APIs like the Anthropic API and financial data APIs. Monitor your usage and set limits to avoid unexpected charges. * **Rate Limiting:** APIs often have rate limits (e.g., a maximum number of requests per minute). Implement error handling and retry mechanisms to deal with rate limiting. * **Data Accuracy:** The accuracy of Claude's analysis depends on the quality of the data you provide. Verify the data from your sources. * **Security:** Protect your API keys and other sensitive information. Store them securely (e.g., in environment variables) and avoid hardcoding them in your code. * **Prompt Engineering is Key:** Experiment with different prompts to get the best results from Claude. Iterate and refine your prompts based on the responses you receive. * **Claude's Limitations:** Claude is a powerful language model, but it's not perfect. It can sometimes make mistakes or provide inaccurate information. Always critically evaluate its output. * **Legal and Ethical Considerations:** Be mindful of data privacy regulations and ethical considerations when using company information and stock prices. Ensure you have the right to use the data and that you are not violating any laws or regulations. **Translation to Portuguese (of the key concepts):** * **Monitoring and Control Platform:** Plataforma de Monitoramento e Controle * **Anthropic Claude AI model:** Modelo de IA Claude da Anthropic * **API:** API (Interface de Programação de Aplicações) * **API Key:** Chave de API * **Prompt Engineering:** Engenharia de Prompt (ou Criação de Instruções) * **Data Cleaning:** Limpeza de Dados * **Data Aggregation:** Agregação de Dados * **Rate Limiting:** Limitação de Taxa * **Data Visualization:** Visualização de Dados * **Alerting:** Alertas **Example Portuguese Prompt:** ``` Analise as seguintes informações sobre a [Nome da Empresa]: Descrição da Empresa: [Descrição da Empresa] Indústria: [Indústria] Número de Funcionários: [Número de Funcionários] Preço Atual das Ações: [Preço Atual das Ações] Preço de Fechamento do Dia Anterior: [Preço de Fechamento do Dia Anterior] Forneça um breve resumo da empresa, sua situação financeira atual com base no preço das ações e potenciais riscos ou oportunidades. Concentre-se em insights importantes que seriam relevantes para um analista de negócios. Mantenha a resposta concisa (abaixo de 200 palavras). ``` **In summary, this is a complex integration that requires careful planning, coding, and testing. Start with a small, well-defined use case and gradually expand as you gain experience.** Remember to consult the documentation for your MCP and the Anthropic Claude API for the most up-to-date information. Good luck!

Genai Toolbox

Servidor MCP de código aberto especializado em ferramentas fáceis, rápidas e seguras para bancos de dados.

mcp-test-server

A lightweight MCP test server for verifying client connectivity, providing tools, resources, and prompts for integration.

MCP Server

A server implementing Model Coupling Protocol for HDF5 file operations, Slurm job management, hardware monitoring, and data compression.

cite-mcp

Recupere dados de citação sem esforço do CiteAs e do Google Scholar. Obtenha citações formatadas em BibTeX para seus recursos com apenas alguns comandos. Melhore seu fluxo de trabalho de pesquisa integrando a recuperação de citações diretamente em seus aplicativos.

Tesla MCP Server

Enables control and monitoring of Tesla vehicles through the Tessie API, providing access to vehicle state (battery, location, climate) and commands (lock/unlock, climate control, navigation) with support for both local and Cloudflare Worker deployment.

CVE MCP Server

Provides conversational access to a local CVE (Common Vulnerabilities and Exposures) database, enabling natural language queries to search vulnerabilities, retrieve detailed CVE information, and view security statistics.

AlibabaCloud DevOps MCP Server

Enables AI assistants to interact with Alibaba Cloud Yunxiao DevOps platform for managing projects, code repositories, work items, pipelines, deployments, and testing workflows through comprehensive organization, development, and delivery tools.

Scripting Docs MCP

Provides semantic search over ScriptingApp documentation by converting Markdown/MDX files into a LlamaIndex vector store. Supports multi-language indexing and enables natural language queries against technical documentation through MCP tools.

News MCP Server for VnExpress

Enables AI agents to interact with VnExpress news content through RSS feeds, advanced search with filters, and clean article text extraction for Vietnamese news consumption.

Ignition MCP Server

Enables AI assistants to interact with Inductiveautomation's Ignition SCADA/MES platform through its REST API. Provides 45+ tools for gateway management, project operations, backup handling, log analysis, and license activation.

Search Service

O Servidor Search MCP permite a integração perfeita de recursos de pesquisa de rede e local em ferramentas como Claude Desktop e Cursor, utilizando a API Brave Search para solicitações assíncronas e de alta simultaneidade.

Figma MCP PRO

Professional Model Context Protocol server that enables AI-optimized Figma design analysis and comprehensive design-to-code conversion through a structured 5-step workflow.

Medium MCP Server

Enables AI assistants to interact with Medium's platform for publishing, updating, and managing articles and drafts through OAuth 2.0 authentication with automatic retry logic and rate limit handling.

DNDzgz MCP Server

An MCP server that provides real-time information about the Zaragoza tram system, including arrival estimations and station details through the DNDzgz API.

SAP Documentation MCP Server

Provides offline access to SAP documentation and real-time SAP Community content, integrating official documentation with community-driven solutions for comprehensive developer support.

AI Cognitive Nexus

Enables dynamic creation and orchestration of hierarchical AI agent teams with role-based personas and domain knowledge injection. Supports multi-agent collaboration, session management, and complex task execution through structured team workflows.

WSB Analyst MCP Server

A Model Context Protocol server that enables LLM clients to fetch, analyze, and extract insights from real-time WallStreetBets posts, comments, and shared links for market analysis.

Azure Container Apps Weather MCP Server

A server-sent events (SSE) MCP server that runs on Azure Container Apps with API key authentication, likely providing weather-related functionality based on the configuration.

Seamless Sign-ups MCP Server

A demonstration project that uses Google Gemini 2.0 Flash to interact with a locally hosted Model Calling Protocol server for managing user registration data stored in CSV files.

Overleaf MCP Server

Enables read-only interaction with Overleaf LaTeX projects through compatible clients like Claude Desktop, Cursor, and VS Code. Allows users to list and read project files safely without modification capabilities.

bilibili_mcp_server_full

Servidor MCP do Bilibili de fácil instalação.