MCP Servers

20 open source tools compared. Sorted by stars. Scroll down for our analysis.

Tool	Stars	Velocity	Language	License	Score
context7 Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors	53.0k	—	TypeScript	—	60
fastmcp The fast, Pythonic way to build MCP servers and clients	24.6k	—	Python	—	77
mcp-use Fullstack MCP framework for building AI-powered apps with interactive widgets	9.8k	—	TypeScript	—	69
amazon-bedrock-agentcore-mcp-server Model Context Protocol (MCP) server for Amazon Bedrock AgentCore services	8.8k	—		—	53
mcp-integration-with-nova-canvas This repository outlines a basic implementation of the [Model Context Protocol](https://modelcontextprotocol.io/) integration with Amazon Nova Canvas for image generation	8.8k	—		—	53
redshift-mcp-server Model Context Protocol (MCP) server for Amazon Redshift.	8.8k	—		—	53
document-loader-mcp-server Model Context Protocol (MCP) server for document parsing and content extraction	8.8k	—		—	53
finch-mcp-server A Model Context Protocol (MCP) server for Finch that enables generative AI models to build and push container images through finch cli leveraged MCP tools.	8.8k	—		—	53
amazon-mq-mcp-server A Model Context Protocol (MCP) server for Amazon MQ that enables generative AI models to manage RabbitMQ and ActiveMQ message brokers through MCP tools.	8.8k	—		—	53
mcp-integration-with-kb This repository outlines a basic implementation of the [Model Context Protocol](https://modelcontextprotocol.io/) integration with Amazon Bedrock Knowledge Bases	8.8k	—		—	53
bedrock-kb-retrieval-mcp-server MCP server for accessing Amazon Bedrock Knowledge Bases	8.8k	—		—	53
sagemaker-ai-mcp-server The Amazon SageMaker AI MCP server provides agents with tools to enable high-performance, low-cost AI/ML model development. Currently, this server includes tools for managing SageMaker HyperPod cluste	8.8k	—		—	53
openapi-mcp-server This project is a server that dynamically creates Model Context Protocol (MCP) tools and resources from OpenAPI specifications. It allows Large Language Models (LLMs) to interact with APIs through the	8.8k	—		—	53
registry Community-driven registry for discovering and publishing MCP servers	6.7k	—	Go	—	67
5ire Cross-platform desktop AI assistant with MCP support and local knowledge bases	5.2k	—	TypeScript	—	65
go-sdk Official Go SDK for building MCP servers and clients	4.4k	—	Go	—	63
shadcn-ui-mcp-server A mcp server to allow LLMS gain context about shadcn ui component structure,usage and installation,compaitable with react,svelte 5,vue & React Native	2.7k	—	TypeScript	—	47
markdownify-mcp A Model Context Protocol server for converting almost anything to Markdown	2.6k	—	TypeScript	—	47
arxiv-mcp-server A Model Context Protocol server for searching and analyzing arXiv papers	2.6k	—	Python	—	47
brightdata-mcp A powerful Model Context Protocol (MCP) server that provides an all-in-one solution for public web access.	2.3k	—	JavaScript	—	47

Stay ahead of the category

New tools and momentum shifts, every Wednesday.

Our Analysis

context753.0k★

Context7 feeds your AI assistant up-to-date documentation for libraries and frameworks. Instead of your LLM guessing API signatures from stale training data, it pulls current docs on demand. One config entry, no API key. Maintained by Upstash and covers a wide range of popular libraries. The docs are pre-processed and chunked for LLM consumption, so you get relevant sections rather than entire documentation sites. Setup is about as simple as MCP gets. Install this. It makes every AI coding session better by grounding responses in real documentation. The catch: coverage depends on what Context7 has indexed. Niche or brand-new libraries may not be available yet.

fastmcp24.6k★

FastMCP is the easiest way to build MCP servers in Python. Decorate a function, and it becomes a tool your AI assistant can call. Schema generation, validation, auth, transport: all handled. Setup is one pip install and a few lines of code. If you've written a Flask route, you can write an MCP server. Prefect maintains this actively and it's now the foundation of the official MCP Python SDK. Free, Apache 2.0. No underlying service costs unless you opt into Prefect's hosted platform. The standard choice for Python MCP development.

mcp-use9.8k★

Framework for building full MCP applications with interactive React widgets. Works across ChatGPT, Claude, and other MCP-compatible clients. Think of it as a starter kit for MCP app development. Scaffold a new project with one npx command. Includes 15+ templates, a web debugger, and multi-language support. TypeScript and Python both work. More opinionated than FastMCP, more focused on end-user apps than raw server tooling. Maintained by the Manufact team. Free and MIT licensed. Optional paid cloud hosting available but not required.

amazon-bedrock-agentcore-mcp-server8.8k★

Bedrock AgentCore MCP connects your AI assistant to Amazon Bedrock's agent orchestration layer. Manage agents, invoke actions, and inspect agent configurations. Setup requires Bedrock access and AgentCore resources provisioned. Bedrock AgentCore pricing includes per-invocation and per-step charges. The MCP is free. AWS Labs maintains it. Useful for teams building multi-step AI agents on Bedrock who want to manage them from their editor instead of the console. The catch: AgentCore is relatively new. The API surface is still evolving, so expect breaking changes as the service matures.

mcp-integration-with-nova-canvas8.8k★

Nova Canvas MCP connects your AI assistant to Amazon Nova Canvas, Bedrock's image generation model. Generate, edit, and manipulate images through text prompts from your editor. Setup requires Bedrock access with Nova Canvas model enabled. Bedrock charges per image generated. Pricing depends on resolution and model version. The MCP is free. AWS Labs maintains it as a reference integration. Interesting for prototyping visual content without leaving your workflow. The catch: like the Knowledge Base integration, this is reference code. Production use requires hardening around error handling and rate limits.

redshift-mcp-server8.8k★

Redshift MCP connects your AI assistant to Amazon Redshift, AWS's data warehouse. Query tables, explore schemas, and analyze data directly from your editor. Config is your Redshift connection details plus AWS credentials. Redshift pricing varies by cluster type. Serverless bills per query. The MCP is free. AWS Labs maintains it. Strong pick for data teams already on Redshift who want AI-assisted analysis without context-switching. The catch: Redshift queries can get expensive fast. No built-in guardrails on query cost, so watch your serverless RPU usage.

document-loader-mcp-server8.8k★

Loads documents into your AI assistant's context. Supports PDF, HTML, Markdown, CSV, Excel, and plain text. You can point it at local files or URLs and get the content extracted and ready for analysis or summarization. The MCP server is free and open source with no underlying service costs. It runs entirely locally using Python-based document parsing. Setup is installing the server and its dependencies. No AWS account needed. Maintained by AWS Labs, but this one is AWS-agnostic. Handy utility for anyone who regularly feeds documents into AI conversations. Simple, does one thing well. Worth installing if you work with varied document formats.

finch-mcp-server8.8k★

Connects your AI assistant to Finch, AWS's open source container development tool (an alternative to Docker Desktop). You can build images, manage containers, and run compose stacks. Think of it as Docker commands you can describe in natural language. The MCP server is free and open source. Finch itself is free and open source with no licensing concerns, unlike Docker Desktop's per-seat pricing for larger companies. Setup requires Finch installed on your machine. Maintained by AWS Labs. Good fit if you've already switched from Docker Desktop to Finch or are evaluating the move. The MCP integration makes container operations conversational. If you're happy with Docker Desktop, no reason to switch just for this.

amazon-mq-mcp-server8.8k★

Connects your AI assistant to Amazon MQ, which runs managed ActiveMQ and RabbitMQ brokers. Manage brokers, inspect queues, publish messages, and troubleshoot message flow without switching to the AWS console. Requires AWS credentials and an existing Amazon MQ broker. The MCP server is free. Amazon MQ pricing is per-hour based on broker instance size, starting around $0.03/hour for small instances. Maintained by AWS Labs. Solid pick if you manage message brokers on AWS and want AI-assisted operations. Not worth setting up Amazon MQ just to use this server.

mcp-integration-with-kb8.8k★

This integration connects your AI assistant to Amazon Bedrock Knowledge Bases. Query your ingested documents using RAG (retrieval-augmented generation) directly from your editor. Setup requires a Bedrock Knowledge Base already configured with data sources. Bedrock Knowledge Base pricing includes storage, indexing, and retrieval charges. The MCP is free. AWS Labs maintains it as a reference integration. Powerful if you have company docs indexed in Bedrock and want your AI assistant to search them. The catch: this is a reference implementation, not a production-ready server. Expect to customize it for your use case.

bedrock-kb-retrieval-mcp-server8.8k★

Connects your AI assistant to Amazon Bedrock Knowledge Bases for retrieval-augmented generation. You can query your indexed documents, get relevant passages with source citations, and use your company's knowledge base directly in conversation. The MCP server is free and open source. Bedrock Knowledge Bases charges for the underlying embedding model, vector store (OpenSearch Serverless or Aurora), and retrieval queries. Costs vary widely based on corpus size. Setup requires an existing Knowledge Base ID and proper IAM permissions. Maintained by AWS Labs. If you've already invested in Bedrock Knowledge Bases, this is the natural way to wire them into your AI workflow. Not useful until you have a knowledge base configured and populated.

sagemaker-ai-mcp-server8.8k★

SageMaker MCP connects your AI assistant to Amazon SageMaker's ML platform. Manage training jobs, endpoints, and model artifacts from your editor. Setup requires AWS credentials and SageMaker resources already provisioned. SageMaker pricing is complex: compute by the hour, storage by the GB, endpoints by uptime. The MCP itself is free. AWS Labs maintains it. Useful for ML engineers who live in SageMaker and want to check job status or deploy models without the console. The catch: SageMaker's API surface is enormous. The MCP exposes a subset, so you will still hit the console for advanced workflows.

openapi-mcp-server8.8k★

OpenAPI MCP takes any OpenAPI spec and turns it into callable MCP tools automatically. Point it at a spec file or URL and your AI assistant can call every endpoint. No hand-written tool definitions needed. Completely free. The underlying APIs you call may have their own costs, but the MCP itself is just a translation layer. AWS Labs maintains it. This is one of the most versatile servers in the AWS MCP suite. Any API with an OpenAPI spec becomes an AI tool instantly. The catch: auto-generated tools can be noisy. Large specs produce dozens of tools, and your AI assistant may struggle to pick the right one.

registry6.7k★

Community-driven registry for discovering and publishing MCP servers. Search for servers by capability, install them into your AI workflow, or publish your own with verified ownership. Built in Go, runs via Docker or standalone binary. Publishing requires identity verification through GitHub OAuth, DNS, or HTTP challenges. The registry API is stable at v0.1 with a CLI for submissions. Maintained by the Model Context Protocol organization (Anthropic-backed). Free, open source. This is the central directory for the MCP ecosystem, worth bookmarking if you build or consume MCP servers.

5ire5.2k★

Desktop AI assistant that connects to multiple LLM providers and supports MCP servers natively. Load local documents (PDFs, spreadsheets, code) into a knowledge base and query them with any model. Cross-platform (Mac, Windows, Linux). Setup requires Python + Node.js + the uv package manager, so it's not a one-click install. Built-in MCP marketplace for adding server integrations. Community maintained. Free, open source. More of an MCP client than a server, but useful if you want a local AI interface that plugs into the MCP ecosystem.

go-sdk4.4k★

Official Go SDK for building MCP servers and clients. Handles the JSON-RPC transport, tool registration, resource management, and OAuth primitives so you write the business logic. Standard Go module install. Well-documented with example implementations. Supports the full MCP specification including streaming and sampling. Maintained by the Model Context Protocol organization. Free, open source. If you're building MCP servers in Go, this is the canonical choice. Equivalent to FastMCP for Python but for the Go ecosystem.

shadcn-ui-mcp-server2.7k★

This MCP server feeds your AI assistant the full context of shadcn/ui components: structure, props, usage patterns, installation commands. Instead of the LLM hallucinating component APIs from training data, it pulls the real docs. Community-built and focused on one job: making AI-generated UI code actually match current shadcn/ui patterns. Setup is a single config entry. No API keys, no auth, no moving parts. Install this if you use shadcn/ui with an AI coding assistant. The difference in code quality is noticeable, fewer wrong prop names, fewer outdated patterns. The catch: only covers shadcn/ui. If you need broader component library context, you will need additional servers.

markdownify-mcp2.6k★

Markdownify converts URLs, PDFs, and other documents into clean Markdown that your AI assistant can actually read. Paste a URL, get structured text back. No copy-pasting, no manual cleanup. Handles web pages, PDFs, Word docs, and more. The conversion strips navigation, ads, and boilerplate so your assistant gets the content without the noise. Setup is straightforward, just the config entry. No API keys required. Worth installing for anyone who feeds documents to AI assistants regularly. Saves real time versus manual copy-paste workflows. The catch: conversion quality varies with source formatting. Complex layouts, heavy JavaScript pages, and scanned PDFs can produce messy output.

arxiv-mcp-server2.6k★

This MCP server connects your AI assistant to arXiv, the open research paper repository. Search for papers, pull abstracts, download full PDFs, and have your assistant summarize findings without leaving your editor. Community-maintained and built on arXiv's free API. Setup is minimal, just add the config. No API keys needed since arXiv's API is open. The search covers all arXiv categories, and you can filter by date, author, or subject. Install this if you read research papers as part of your work. Having your assistant pull and summarize papers on demand is a real workflow improvement. The catch: arXiv's API has rate limits, and full-text extraction from PDFs is not always clean.

brightdata-mcp2.3k★

Bright Data's MCP server gives your AI assistant access to their web scraping infrastructure: structured data extraction, SERP results, and page rendering through residential proxies. Your assistant asks for data, Bright Data handles the anti-bot circumvention. Official, maintained by Bright Data. Setup requires an API key and credits on your Bright Data account. The MCP itself is free, but every request burns credits. Pricing is usage-based and adds up fast on high-volume scraping. Install this if you already pay for Bright Data and want AI-driven scraping. Do not install it to "try web scraping" because the credit costs will surprise you. The catch: this is a gateway to a paid API. The MCP is free, the data is not.