Guide · Open source
How to use awesome-mcp-servers: browse and pick an MCP server
Updated · written with AI from the project's own documentation and checked against our catalogue data
awesome-mcp-servers is a categorized, community-maintained list of servers that follow the Model Context Protocol (MCP), an open protocol that lets AI models reach files, databases and APIs through standardized server implementations. It is an index that points to other projects, not a single app. This guide shows you how to read the list, use its icons to filter options, and find the next step for the server you choose.
Awesome MCP Serverspunkpeye/awesome-mcp-servers · ★ 97.2k · MIT Install options & downloads Quick start
- 1Open the list. Go to the awesome-mcp-servers README, or clone it with the optional git command.
- 2Check the scope. If you need a hosted server reached by URL, go to awesome-remote-mcp-servers instead.
- 3Pick a category. Use the category index to jump to a section such as Databases or Browser Automation.
- 4Filter by icons. Use the language, scope and operating system icons to narrow entries to what fits your setup.
- 5Open the repository. Follow the entry's link to its repository and read its description.
- 6Install from that project. Follow the install instructions in the chosen project's own README, since this list does not provide them.
What is awesome-mcp-servers?
The README describes the project as a curated list of awesome Model Context Protocol (MCP) servers. MCP is an open protocol that enables AI models to interact with local and remote resources through standardized server implementations.
The list is not a program you run. Each entry links to a repository, and the project it links to is the one you install and use.
- It is an index of other projects, so there is nothing to install from this repository itself.
- It is licensed under MIT.
- The list covers both production-ready and experimental servers, according to the README.
Know which kind of server you need
The list is for servers with a GitHub repo that you install and run yourself. The README states that hosted servers you connect to over a URL are covered by a separate list, awesome-remote-mcp-servers.
If you are looking for a client application that connects to MCP servers, the README points to awesome-mcp-clients.
- Install and run yourself: stay in awesome-mcp-servers.
- Hosted server reached by URL: use awesome-remote-mcp-servers.
- Client apps: use awesome-mcp-clients.
Read the categories and icons
Servers are grouped by category, such as Browser Automation, Databases, File Systems, Developer Tools and Version Control. The category index near the top of the README links to each section.
Each entry uses icons. The Legend section defines them. Language icons show the codebase language, scope icons show cloud or local service, and operating system icons show platform support.
- Language: 🐍 Python, 📇 TypeScript or JavaScript, 🏎️ Go, 🦀 Rust, ☕ Java, 💎 Ruby and others listed in the Legend.
- Scope: ☁️ cloud service, 🏠 local service, 📟 embedded systems.
- Operating system: 🍎 macOS, 🪟 Windows, 🐧 Linux.
- 🎖️ marks an official implementation.
Choose local or cloud
The README gives a simple rule. Use local when the MCP server talks to software installed on your machine, for example controlling a Chrome browser. Use cloud when the server talks to remote APIs, for example a weather API.
Read an entry and follow its project's instructions
Each entry has a name that links to the repository, a short description and icons. The description tells you what the server does. The repository then tells you how to install and configure it.
Installing any entry means following that project's own instructions. This list does not provide a universal install process.
- Open the repository link.
- Check the description and icons for language, scope and operating system.
- Follow the install and configuration steps in that project's README.
- The README also links to a web-based directory at glama.ai/mcp/servers that is synced with the repository.
Optionally clone the list to your computer
You can read the list online without installing anything. If you want a local copy, the README gives this command. The clone step is optional.
git clone https://github.com/punkpeye/awesome-mcp-serversWhat people use it for
- Finding a server for a specific service such as a database or a browser.
- Exploring what kinds of tools AI assistants can be connected to.
- Comparing options in one category, such as browser automation or databases, before building your own server.
Frequently asked questions
Is awesome-mcp-servers a single app I can install?
No. The README describes it as a curated list of MCP servers that links to other projects. Installing an entry means following that specific project's own instructions.
Does awesome-mcp-servers have any prerequisites?
The install method is listed as read-online with no prerequisites. You can browse the README without installing anything. The optional clone uses git.
What is the license?
The project is licensed under MIT, as listed in its metadata.
What is the difference between awesome-mcp-servers and awesome-remote-mcp-servers?
awesome-mcp-servers lists servers you install and run yourself from a GitHub repo. The README says awesome-remote-mcp-servers covers hosted servers you connect to over a URL.
What do the icons mean?
The Legend section defines them. Language icons show the codebase language, 🏠 and ☁️ show local or cloud scope, and 🍎, 🪟 and 🐧 show supported operating systems. 🎖️ marks an official implementation.
Where can I find a web version of the directory?
The README links to a web-based directory at glama.ai/mcp/servers that is synced with the repository.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use awesome-mcp-servers if you are a developer or a team looking for an existing MCP server for a specific service and want to compare options by category. Pick something else if you want one ready-to-run app or a hosted server you connect to by URL, and look at awesome-remote-mcp-servers for the latter.
Install options & downloads