Today, we're excited to announce the Awesome Copilot MCP Server - a simple way to search those customizations and savethem directly to your repository from the MCP server.In early July, we announced the community-driven Awesome GitHubCopilot Customizations repository. It contains chat modes, instructions and prompts that let you tailor the AIresponses from GitHub Copilot, covering tons of situations. The problem is, with so many great options, discovering andcomparing the right ones is hard.In this blog post, I'm going to show you a way to discover these incredibly useful chat modes, instructions and promptsjust by chatting with GitHub Copilot, together with this Awesome Copilot MCP server.What are chat modes, instructions and prompts? Custom chat modes define how chat operates, which tools it can use, and how it interacts withthe codebase. Each chat prompt is run within the boundaries of the chat mode, without having to configure toolsand instructions for every request. Through these custom chat modes, you can create a front-end developer chatmode, where the AI can only generate and modify code related to the front-end app development. For a concreteexample, see the chat mode file examplein the VS Code docs. Custom instructions define common guidelines or rules for tasks like generating code,performing code reviews, or generating commit messages. Custom instructions describe the conditions in which theAI should perform operations - how a task should be done. Through these custominstructions, you can specify coding practices, preferred technologies or project requirements so that thegenerated code follows your standards. Prompts define reusable prompts for common tasks like generating code or performing a codereview. Prompts are standalone prompts that you can run directly in chat. They describe the task to be performed- what should be done. Optionally, you can include task-specific guidelines about howthe task should be performed, or you can reference custom instructions in the prompt file. Through these customprompts, you can create reusable prompts for common coding tasks like scaffolding a new component, API route orgenerating tests.If you're new to these, see the VS Code docs for chat modes, custominstructions, and prompt files.There are hundreds of those custom chat modes, instructions and prompts in the Awesome Copilot repository, and the MCPserver helps you find the best one. Let's walk through how to install the MCP server in VS Code and start using it.Install Awesome Copilot MCP Server[alert type="important" heading="Prerequisite"]Installation requires Docker Desktop installed and running, since thisserver runs in a container.[/alert]Installing this MCP server is really easy. Click the following button to directly install it in VS Code. Alternatively, add the following to your MCP server configuration:{ "servers": { "awesome-copilot": { "type": "stdio", "command": "docker", "args": [ "run", "-i", "--rm", "ghcr.io/microsoft/mcp-dotnet-samples/awesome-copilot:latest" ] } }}There are other ways to install this MCP server. Find more about it at this GitHub repository.Use Awesome Copilot MCP Server[video src="https://devblogs.microsoft.com/wp-content/uploads/2025/08/Awesome.Copilot.MCP_.Edit_.webm" width="1000"height="563"]Demo: Searching and saving Copilot customizations in GitHub Copilot Chat. Portions of this video are sped up forillustration.This MCP server provides two tools and one prompt. Tools #search_instructions: searches GitHub Copilot customizations based on keywords provided. #load_instruction: loads a specific GitHub Copilot customization. Prompts /mcp.awesome-copilot.get_search_prompt: provides a prompt to search GitHub Copilotcustomizations.After installing and running this MCP server, call the prompt in the GitHub Copilot Chat window like:/mcp.awesome-copilot.get_search_promptYou'll be prompted to enter keywords to search.Let's enter "python" as a keyword for example. Then, the following prompt appears in the GitHub Copilot Chat window.Please search all the chatmodes, instructions and prompts that are related to the search keyword, `python`.Here's the process to follow:1. Use the `awesome-copilot` MCP server.2. Search all chatmodes, instructions, and prompts for the keyword provided.3. DO NOT load any chatmodes, instructions, or prompts from the MCP server until the user asks to do so.4. Scan local chatmodes, instructions, and prompts markdown files in `.github/chatmodes`, `.github/instructions`, and `.github/prompts` directories respectively.5. Compare existing chatmodes, instructions, and prompts with the search results.6. Provide a structured response in a table format that includes the already exists, mode (chatmodes, instructions or prompts), filename, title and description of each item found. Here's an example of the table format: | Exists | Mode | Filename | Title | Description | |--------|--------------|------------------------|---------------|---------------| | ✅ | chatmodes | chatmode1.json | ChatMode 1 | Description 1 | | ❌ | instructions | instruction1.json | Instruction 1 | Description 1 | | ✅ | prompts | prompt1.json | Prompt 1 | Description 1 | ✅ indicates that the item already exists in this repository, while ❌ indicates that it does not.7. If any item doesn't exist in the repository, ask which item the user wants to save.8. If the user wants to save it, save the item in the appropriate directory (`.github/chatmodes`, `.github/instructions`, or `.github/prompts`) using the mode and filename, with NO modification.[alert type="tip" heading="Tip"]If you find a suitable customization, reply with the filename (for example,python.instructions.md) to load and save it.[/alert]Send this prompt, and you'll see GitHub Copilot try to connect to the MCP server.The search result in the table format defined in the prompt is now coming up. And at the end of the answer, it askswhich customization to save. If you find a suitable one for your purpose, enter the filename. As you received the listof chat modes, instructions and prompts, just enter one of those files (for example,python.instructions.md).Then, it will load the content via the MCP server.After the content loads, it's saved to .github/instructions/python.instructions.md. If you want to saveanother file, enter the filename again.Try it yourself!Now, it's your turn! Install this MCP server on your local machine, search for GitHub Copilot customization through theserver and save it!More resourcesIf you'd like to learn more about MCP in .NET, here are some additional resources worth exploring: Awesome Copilot Awesome Copilot MCP Server Let's Learn MCP MCP Workshop in .NET MCP Samples in .NET MCP Samples