project-setup-info-local

作者: microsoft

提供關於如何設定各類專案的資訊。適用於根據需求初始化專案及搭建專案架構。

npx skills add https://github.com/microsoft/vscode-copilot-chat --skill project-setup-info-local

How to setup a project

Determine what kind of project the user wants to create, then based on that, choose which setup info below to follow. Only setup a project if the folder is empty or if you've just called the tool first calling the tool to create a workspace.

vscode-extension

A template for creating a VS Code extension using Yeoman and Generator-Code.

Run this command:

npx --package yo --package generator-code -- yo code . --skipOpen

The command has the following arguments:

  • -t, --extensionType: Specify extension type: ts, js, command-ts, command-js, colortheme, language, snippets, keymap, extensionpack, localization, commandweb, notebook. Defaults to ts
  • -n, --extensionDisplayName: Set the display name of the extension.
  • --extensionId: Set the unique ID of the extension. Do not select this option if the user has not requested a unique ID.
  • --extensionDescription: Provide a description for the extension.
  • --pkgManager: Specify package manager: npm, yarn, or pnpm. Defaults to npm.
  • --bundler: Bundle the extension using webpack or esbuild.
  • --gitInit: Initialize a Git repository for the extension.
  • --snippetFolder: Specify the location of the snippet folder.
  • --snippetLanguage: Set the language for snippets.

Rules

  1. Do not remove any arguments from the command. Only add arguments if the user requests them.
  2. Call the tool get_vscode_api with the user's query to get the relevant references.
  3. After the tool get_vscode_api has completed, only then begin to modify the project.

next-js

A React based framework for building server-rendered web applications.

Run this command:

npx create-next-app@latest .

The command has the following arguments:

  • --ts, --typescript: Initialize as a TypeScript project. This is the default.
  • --js, --javascript: Initialize as a JavaScript project.
  • --tailwind: Initialize with Tailwind CSS config. This is the default.
  • --eslint: Initialize with ESLint config.
  • --app: Initialize as an App Router project.
  • --src-dir: Initialize inside a 'src/' directory.
  • --turbopack: Enable Turbopack by default for development.
  • --import-alias <prefix/*>: Specify import alias to use.(default is "@/*")
  • --api: Initialize a headless API using the App Router.
  • --empty: Initialize an empty project.
  • --use-npm: Explicitly tell the CLI to bootstrap the application using npm.
  • --use-pnpm: Explicitly tell the CLI to bootstrap the application using pnpm.
  • --use-yarn: Explicitly tell the CLI to bootstrap the application using Yarn.
  • --use-bun: Explicitly tell the CLI to bootstrap the application using Bun.

vite

A front end build tool for web applications that focuses on speed and performance. Can be used with React, Vue, Preact, Lit, Svelte, Solid, and Qwik.

Run this command:

npx create-vite@latest .

The command has the following arguments:

  • -t, --template NAME: Use a specific template. Available templates: vanilla-ts, vanilla, vue-ts, vue, react-ts, react, react-swc-ts, react-swc, preact-ts, preact, lit-ts, lit, svelte-ts, svelte, solid-ts, solid, qwik-ts, qwik

mcp-server

A Model Context Protocol (MCP) server project. This project supports multiple programming languages including TypeScript, JavaScript, Python, C#, Java, and Kotlin.

Rules

  1. First, visit https://github.com/modelcontextprotocol to find the correct SDK and setup instructions for the requested language. Default to TypeScript if no language is specified.
  2. Use the fetch_webpage tool to find the correct implementation instructions from https://modelcontextprotocol.io/llms-full.txt
  3. Update the copilot-instructions.md file in the .github directory to include references to the SDK documentation
  4. Create an mcp.json file in the .vscode folder in the project root with the following content: { "servers": { "mcp-server-name": { "type": "stdio", "command": "command-to-run", "args": [list-of-args] } } }.
    • mcp-server-name: The name of the MCP server. Create a unique name that reflects what this MCP server does.
    • command-to-run: The command to run to start the MCP server. This is the command you would use to run the project you just created.
    • list-of-args: The arguments to pass to the command. This is the list of arguments you would use to run the project you just created.
  5. Install any required VS Code extensions based on the chosen language (e.g., Python extension for Python projects).
  6. Inform the user that they can now debug this MCP server using VS Code.

python-script

A simple Python script project which should be chosen when just a single script wants to be created.

Required extensions: ms-python.python, ms-python.vscode-python-envs

Rules

  1. Call the tool copilot_runVscodeCommand to correctly create a new Python script project in VS Code. Call the command with the following arguments. Note that "python-script" and "true" are constants while "New Project Name" and "/path/to/new/project" are placeholders for the project name and path respectively.
    {
      "name": "python-envs.createNewProjectFromTemplate",
      "commandId": "python-envs.createNewProjectFromTemplate",
      "args": ["python-script", "true", "New Project Name", "/path/to/new/project"]
    }
    

python-package

A Python package project which can be used to create a distributable package.

Required extensions: ms-python.python, ms-python.vscode-python-envs

Rules

  1. Call the tool run_vscode_command to correctly create a new Python package project in VS Code. Call the command with the following arguments. Note that "python-package" and "true" are constants while "New Package Name" and "/path/to/new/project" are placeholders for the package name and path respectively.
    {
      "name": "python-envs.createNewProjectFromTemplate",
      "commandId": "python-envs.createNewProjectFromTemplate",
      "args": ["python-package", "true", "New Package Name", "/path/to/new/project"]
    }
    

來自 microsoft 的更多技能

oss-growth
microsoft
開源增長駭客角色
agent-framework-azure-ai-py
microsoft
使用Microsoft Agent Framework Python SDK(agent-framework-azure-ai)构建Azure AI Foundry代理。适用于使用AzureAIAgentsProvider创建持久化代理、使用托管工具(代码解释器、文件搜索、网络搜索)、集成MCP服务器、管理对话线程或实现流式响应。涵盖函数工具、结构化输出和多工具代理。
development
airunway-aks-setup
microsoft
在AKS上設定AI Runway——從裸叢集到執行模型。涵蓋叢集驗證、控制器安裝、GPU評估、供應商設定及首次部署。時機:「設定AI Runway」、「上線AKS叢集」、「安裝AI Runway」、「airunway設定」、「部署模型至AKS」、「在AKS上進行GPU推論」、「在AKS上設定KAITO」、「在AKS上執行LLM」、「在AKS上使用vLLM」、「在AKS上設定模型服務」、「AI Runway控制器」。
devops
appinsights-instrumentation
microsoft
使用Azure Application Insights檢測Web應用程式的指南。提供遙測模式、SDK設定與組態參考。適用時機:如何檢測應用程式、App Insights SDK、遙測模式、什麼是App Insights、Application Insights指南、檢測範例、APM最佳實踐。
devops
applicationinsights-web-ts
microsoft
使用Application Insights JavaScript SDK(@microsoft/applicationinsights-web)為瀏覽器/Web應用程式進行檢測。適用於真實使用者監控(RUM)——頁面檢視、點擊、AJAX/fetch依賴、例外、自訂事件,以及與後端OpenTelemetry追蹤關聯的瀏覽器端GenAI代理追蹤。涵蓋SDK載入器指令碼與npm設定、框架擴充(React、React Native、Angular)、點擊分析、遙測初始化器,以及從瀏覽器發出的代理/工具/模型span的OTel GenAI語意慣例。
devops
azure-ai-anomalydetector-java
microsoft
使用適用於 Java 的 Azure AI 異常偵測器 SDK 建置異常偵測應用程式。在實作單變量/多變量異常偵測、時間序列分析或 AI 驅動監控時使用。
development
azure-ai-language-conversations-py
microsoft
使用 azure-ai-language-conversations Python SDK 實作對話語言理解(CLU)。當使用 ConversationAnalysisClient 分析對話意圖與實體、建置 NLP 功能,或將語言理解整合至應用程式時使用。
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python。用於機器學習工作區、作業、模型、資料集、計算資源與管線。 觸發詞:「azure-ai-ml」、「MLClient」、「workspace」、「model registry」、「training jobs」、「datasets」。
development