Discover Awesome MCP Servers

Ramp MCP

Un servidor MCP que se conecta a la API de Ramp, permitiendo a los usuarios interactuar con sus datos financieros de Ramp a través del lenguaje natural utilizando clientes MCP como Cursor o Claude Desktop.

ActionsMCP

A Model Context Protocol server that exposes project-specific development tools (tests, linters, typecheckers) to coding assistants through a simple YAML configuration, eliminating the need for manual command approval.

TVControl

MCP server that lets AI agents directly control and interact with the TradingView desktop app via 88 chart-control tools, enabling automated chart reading, Pine Script compilation, strategy optimization, and replay control.

komodo-mcp

Manage Komodo DevOps platform resources (servers, containers, stacks, builds) directly from AI assistants via MCP protocol. Supports OAuth 2.1 and easy Docker Compose deployment.

Library Access MCP

Enables AI assistants to access authenticated academic databases via browser sessions, eliminating re-login and password exposure.

Gemini MCP

Connects Claude.ai with Google's Gemini API to generate images and videos using your own API key.

Unit Expert MCP

Converts common units for length, weight, temperature, area, and volume.

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.

mcp-common-crawl

MCP server for Common Crawl CDX that enables backlink discovery, expired domain finding, and competitor gap analysis without requiring API keys.

sessionmem

Local-first MCP server that watches your coding sessions and injects a compact summary at the start of each new session. 85.6% token reduction, SQLite storage, no cloud. Works with Claude Code, Cursor, Cline, and Windsurf.

QuickChart MCP Server

Generates and downloads charts using QuickChart.io through Claude Desktop and other MCP clients.

Bayesian MCP

A Model Calling Protocol server that enables LLMs to perform rigorous Bayesian analysis and probabilistic reasoning, including inference, model comparison, and predictive modeling with uncertainty quantification.

@fastly/mcp

Enables assistants to interact with the Fastly API, allowing them to manage services, domains, TLS settings, traffic, dictionaries, ACLs, purge content, and make configuration changes.

UnityInfoMCP

A runtime inspection and automation toolkit that enables MCP clients to interact with live Unity game sessions through a dedicated bridge plugin. It allows users to browse scene hierarchies, inspect component fields, search text elements, and modify game object properties in real-time.

Rag chatbot with a localhost MCP server

Okay, I understand. You're building a Retrieval-Augmented Generation (RAG) based HR chatbot that will provide information about workplace rules, and you're planning to use an MCP (presumably a Message Communication Platform or similar) server. Here's a breakdown of the key components and considerations for this project, along with Spanish translations for common terms: **1. Core Components (Componentes Principales):** * **HR Knowledge Base (Base de Conocimiento de RR.HH.):** This is your collection of workplace rules, policies, and procedures. It needs to be well-organized and easily searchable. * *Spanish:* *Base de Conocimiento de Recursos Humanos* * **Embedding Model (Modelo de Incrustación):** This converts your text data (rules, policies) into numerical representations (embeddings) that capture the semantic meaning. Common choices include Sentence Transformers, OpenAI Embeddings, or Cohere Embeddings. * *Spanish:* *Modelo de Incrustación* or *Modelo de Vectores Semánticos* * **Vector Database (Base de Datos Vectorial):** This stores the embeddings, allowing for efficient similarity searches. Examples include Pinecone, Chroma, Weaviate, or FAISS. * *Spanish:* *Base de Datos Vectorial* * **Retrieval Component (Componente de Recuperación):** This takes the user's question, converts it into an embedding, and searches the vector database for the most relevant documents. * *Spanish:* *Componente de Recuperación* * **Large Language Model (LLM) (Modelo de Lenguaje Grande):** This takes the retrieved documents and the user's question and generates a coherent and informative answer. Examples include GPT-3.5, GPT-4, Llama 2, or other open-source models. * *Spanish:* *Modelo de Lenguaje Grande (LLM)* * **MCP Server Integration (Integración con el Servidor MCP):** This allows the chatbot to interact with users through your chosen messaging platform. You'll need to use the MCP's API to send and receive messages. * *Spanish:* *Integración con el Servidor MCP* **2. Workflow (Flujo de Trabajo):** 1. **User Asks a Question (El Usuario Hace una Pregunta):** The user sends a question to the chatbot via the MCP. * *Spanish:* *El usuario hace una pregunta* 2. **Question Embedding (Incrustación de la Pregunta):** The user's question is converted into an embedding using the same embedding model used for the knowledge base. * *Spanish:* *Incrustación de la pregunta* 3. **Similarity Search (Búsqueda de Similitud):** The embedding is used to search the vector database for the most relevant documents. * *Spanish:* *Búsqueda de Similitud* 4. **Contextualization (Contextualización):** The retrieved documents are combined with the user's question to provide context for the LLM. * *Spanish:* *Contextualización* 5. **Answer Generation (Generación de la Respuesta):** The LLM generates an answer based on the context. * *Spanish:* *Generación de la Respuesta* 6. **Response to User (Respuesta al Usuario):** The chatbot sends the answer back to the user via the MCP. * *Spanish:* *Respuesta al Usuario* **3. Key Considerations (Consideraciones Clave):** * **Data Quality (Calidad de los Datos):** The accuracy and completeness of your HR knowledge base are crucial. Ensure the information is up-to-date and well-written. * *Spanish:* *Calidad de los Datos* * **Embedding Model Choice (Elección del Modelo de Incrustación):** Choose an embedding model that is appropriate for your data and language. Consider models specifically trained for semantic similarity. * *Spanish:* *Elección del Modelo de Incrustación* * **LLM Choice (Elección del LLM):** Consider the cost, performance, and capabilities of different LLMs. Fine-tuning the LLM on your HR data can improve accuracy and relevance. * *Spanish:* *Elección del LLM* * **MCP Integration (Integración con MCP):** Understand the API of your MCP server and how to send and receive messages programmatically. * *Spanish:* *Integración con MCP* * **Security (Seguridad):** Implement appropriate security measures to protect sensitive HR data. * *Spanish:* *Seguridad* * **Scalability (Escalabilidad):** Design your system to handle a large number of users and requests. * *Spanish:* *Escalabilidad* * **User Experience (Experiencia del Usuario):** Make the chatbot easy to use and understand. Provide clear instructions and helpful feedback. * *Spanish:* *Experiencia del Usuario* * **Prompt Engineering (Ingeniería de Prompts):** Craft effective prompts for the LLM to ensure it generates accurate and relevant answers. This includes providing clear instructions and examples. * *Spanish:* *Ingeniería de Prompts* * **Evaluation (Evaluación):** Regularly evaluate the performance of the chatbot and make adjustments as needed. Track metrics such as accuracy, relevance, and user satisfaction. * *Spanish:* *Evaluación* **4. Example Technologies/Libraries (Ejemplos de Tecnologías/Librerías):** * **Python:** A popular language for building chatbots and working with LLMs. * *Spanish:* *Python* * **Langchain:** A framework for building applications powered by LLMs. * *Spanish:* *Langchain* (usually used as is) * **LlamaIndex (GPT Index):** Another framework for building applications powered by LLMs, focused on data indexing and retrieval. * *Spanish:* *LlamaIndex* (usually used as is) * **FastAPI/Flask:** Frameworks for building APIs to connect the chatbot to the MCP server. * *Spanish:* *FastAPI/Flask* (usually used as is) **5. Example Code Snippet (Conceptual - Python with Langchain):** ```python # Conceptual example - requires setup of vectorstore, LLM, and MCP integration from langchain.chains import RetrievalQA from langchain.document_loaders import TextLoader from langchain.embeddings.openai import OpenAIEmbeddings from langchain.llms import OpenAI from langchain.text_splitter import CharacterTextSplitter from langchain.vectorstores import Chroma # 1. Load documents (Cargar documentos) loader = TextLoader("workplace_rules.txt") documents = loader.load() # 2. Split documents (Dividir documentos) text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0) texts = text_splitter.split_documents(documents) # 3. Create embeddings (Crear incrustaciones) embeddings = OpenAIEmbeddings() # 4. Create vectorstore (Crear base de datos vectorial) db = Chroma.from_documents(texts, embeddings) # 5. Create retriever (Crear recuperador) retriever = db.as_retriever() # 6. Create LLM (Crear LLM) llm = OpenAI() # 7. Create QA chain (Crear cadena de preguntas y respuestas) qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever) # 8. Get user question from MCP (Obtener pregunta del usuario desde MCP) user_question = get_user_question_from_mcp() # Replace with your MCP integration # 9. Generate answer (Generar respuesta) answer = qa.run(user_question) # 10. Send answer to user via MCP (Enviar respuesta al usuario a través de MCP) send_answer_to_user_via_mcp(answer) # Replace with your MCP integration print(answer) ``` **Important Notes:** * **MCP Specifics:** The most crucial part is the integration with your MCP server. You'll need to consult the MCP's documentation to understand how to send and receive messages. This will likely involve using their API. * **Error Handling:** Implement robust error handling to gracefully handle unexpected situations. * **Testing:** Thoroughly test your chatbot to ensure it provides accurate and helpful information. * **Iterative Development:** Start with a small set of rules and policies and gradually expand your knowledge base. Continuously evaluate and improve the chatbot's performance. This detailed breakdown should give you a solid foundation for building your RAG-based HR chatbot. Remember to adapt the specific technologies and techniques to your specific needs and resources. Good luck! (¡Buena suerte!)

clawboard-mcp-server

Enables AI agents to interact with ClawBoard tasks, including listing, searching, viewing details, posting comments, and upvoting.

data-olympus MCP server

Provides a single-writer MCP server for a governance-grade knowledge base of markdown documents with version control and query capabilities.

ai-test-gen-poc

An AI-powered test script generator that uses Playwright MCP Server to generate test scripts in TypeScript using BDD-style prompts

Morphik MCP

Enables interaction with the Morphik multi-modal database system for document ingestion, retrieval, querying, and management. Supports text and file ingestion, semantic search with LLM-powered completions, and file system navigation with security controls.

Canvas MCP Server

Connects Canvas LMS to AI assistants, enabling users to list courses and retrieve assignment details through natural language. It features secure multi-institution support with encrypted token storage and a simplified setup process for students.

kusto-mcp

Enables AI assistants to query Azure Data Explorer using natural language, eliminating the need to write KQL.

Lelly MCP Server

Connects Lelly.chat workspace with AI agents, offering tools for tasks, reminders, health tracking, CRM, knowledge base, finance, and spiritual journaling via MCP.

remote-mcp-server-authless

Deploy a remote MCP server on Cloudflare Workers without authentication, enabling use with Cloudflare AI Playground and local clients via mcp-remote proxy.

mcp-gmail

Enables interaction with Gmail through the Gmail API to read unread messages, retrieve email threads, and create draft replies programmatically. Supports OAuth authentication and automated email processing through natural language.

twitterapis

MCP server for the Twitter/X read API, enabling search, user profiles, tweets, followers, and more via natural language.

trademe-tinymcp

A simple MCP server for interacting with Trade Me APIs, providing tools for connectivity testing and SDK integration.

Ultra MCP-SS

A FastAPI-based MCP server that integrates with smartscreen.tv, allowing you to programmatically control web displays by displaying media, sending notifications, and controlling playback via HTTP commands.

Medical Calculator MCP Service

Provides access to 56 professional medical calculators covering multiple specialties including cardiovascular, renal, hepatic, respiratory, and critical care assessments. Enables healthcare professionals to perform evidence-based calculations with detailed explanations and automatic unit conversions.

TimeSeriesMCPServer

时间序列预测MCP服务器，支持RNN、LSTM、GRU三种深度学习模型的训练、对比和预测。

dart-query

An MCP server for Dart AI task management that enables bulk operations using DartQL selectors to minimize token usage and context rot. It provides tools for batch updates, task and document CRUD, and safe CSV imports with integrated dry-run capabilities.