Beispiele

Jede Datei hier liegt in samples/ und wird in CI gegen alle drei Plattformen validiert — nichts auf dieser Seite kann von der Engine abweichen.

Kleinste nützliche Config

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: the smallest useful config - check node, install, start a dev server.
version: 1
name: Vite dev server
requires:
  - tool: node
    version: ">=20"
    install: https://nodejs.org/en/download
setup:
  - npm ci
tasks:
  - name: web
    run: npm run dev
    readyWhen: {port: 5173}
    open: true

.NET-Webanwendung

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: verifying an SDK version before running an ASP.NET Core app.
version: 1
name: ASP.NET Core API
requires:
  - tool: dotnet
    version: ">=9.0"
    install: https://dot.net
env:
  ASPNETCORE_ENVIRONMENT: Development
setup:
  - dotnet restore
tasks:
  - name: api
    run: dotnet run --project src/Api
    readyWhen: {log: 'Now listening on: (?<url>\S+)'}
    open: true

Python mit virtueller Umgebung

Zeigt, wie Platform-Maps mit dem Layout-Unterschied der venv zwischen Windows und dem Rest umgehen.

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: a virtual environment, and how platform maps handle the venv layout difference.
version: 1
name: Flask app
requires:
  - tool: python
    version: ">=3.11"
    install: https://www.python.org/downloads/
setup:
  - run:
      windows: python -m venv .venv
      linux: python3 -m venv .venv
      macos: python3 -m venv .venv
  - run:
      windows: .venv\Scripts\pip install -r requirements.txt
      linux: .venv/bin/pip install -r requirements.txt
      macos: .venv/bin/pip install -r requirements.txt
tasks:
  - name: web
    run:
      windows: .venv\Scripts\python app.py
      linux: .venv/bin/python app.py
      macos: .venv/bin/python app.py
    readyWhen: {port: 5000}
    open: true

Mehrere Dienste gleichzeitig

Postgres, eine .NET-API und ein Vite-Frontend, in Abhängigkeitsreihenfolge gestartet und beim Stoppen aufgeräumt.

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: three runtimes at once, started in dependency order, cleaned up on stop.
version: 1
name: Full stack
requires:
  - tool: docker
    install: https://docs.docker.com/get-docker/
  - tool: dotnet
    version: ">=9.0"
  - tool: node
    version: ">=20"
setup:
  - run: npm ci
    cwd: web
  - dotnet restore
tasks:
  - name: db
    run: docker compose up -d postgres
    readyWhen: {port: 5432}
  - name: api
    run: dotnet run --project src/Api
    dependsOn: [db]
    env:
      ConnectionStrings__Default: Host=localhost;Database=app;Username=postgres;Password=postgres
    readyWhen: {http: "http://localhost:5000/health"}
    restart: onFailure
  - name: web
    run: npm run dev
    cwd: web
    dependsOn: [api]
    readyWhen: {port: 5173}
    open: true
stop:
  - docker compose down

Eine bestehende Compose-Datei einpacken

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: wrapping an existing compose file, including teardown.
version: 1
name: Compose stack
requires:
  - tool: docker
    install: https://docs.docker.com/get-docker/
tasks:
  - name: stack
    run: docker compose up
    readyWhen: {http: "http://localhost:8080"}
    open: true
stop:
  - docker compose down -v

Ein generiertes Eingabeformular

Erforderliches Secret mit validiertem Muster, eine Zahl mit Bereich, ein Dropdown und ein Schalter.

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: a generated form - required secret, validated pattern, number range, dropdown.
version: 1
name: AI chat demo
inputs:
  - id: apiKey
    label: OpenAI API key
    type: password
    description: Created at platform.openai.com
    required: true
    pattern: "^sk-"
    env: OPENAI_API_KEY
  - id: port
    label: Port
    type: number
    default: "3000"
    min: 1024
    max: 65535
    env: PORT
  - id: model
    label: Model
    type: select
    options: [gpt-4o-mini, gpt-4o]
    default: gpt-4o-mini
    env: MODEL
  - id: verbose
    label: Verbose logging
    type: bool
    default: "false"
    env: VERBOSE
setup:
  - npm ci
tasks:
  - name: app
    run: npm start
    readyWhen: {port: 3000}
    open: "http://localhost:${inputs.port}"

Ein Skript pro Plattform

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: one script per platform, the shortest possible config.
version: 1
name: Script runner
run:
  windows: powershell -ExecutionPolicy Bypass -File ./run.ps1
  linux: ./run.sh
  macos: ./run.sh

Ein Repository, das sein SDK mitbringt

Nichts zu require, weil das Setup installiert, was es braucht.

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# What this shows: a repository that brings its own SDK, so there is nothing to require.
version: 1
name: Self-contained .NET sample
setup:
  - run:
      linux: curl -sSL https://dot.net/v1/dotnet-install.sh -o dotnet-install.sh && chmod +x dotnet-install.sh && ./dotnet-install.sh -c 10.0 --install-dir ./dotnet10
      macos: curl -sSL https://dot.net/v1/dotnet-install.sh -o dotnet-install.sh && chmod +x dotnet-install.sh && ./dotnet-install.sh -c 10.0 --install-dir ./dotnet10
      windows: powershell -NoProfile -Command "Invoke-WebRequest https://dot.net/v1/dotnet-install.ps1 -OutFile dotnet-install.ps1; ./dotnet-install.ps1 -Channel 10.0 -InstallDir ./dotnet10"
  - run:
      linux: ./dotnet10/dotnet restore
      macos: ./dotnet10/dotnet restore
      windows: .\dotnet10\dotnet restore
tasks:
  - name: sample
    run:
      linux: ./dotnet10/dotnet run --project ./Samples/MainSample.WebAssembly/MainSample.WebAssembly.csproj
      macos: ./dotnet10/dotnet run --project ./Samples/MainSample.WebAssembly/MainSample.WebAssembly.csproj
      windows: .\dotnet10\dotnet run --project .\Samples\MainSample.WebAssembly\MainSample.WebAssembly.csproj
    readyWhen: {log: 'Now listening on: (?<url>\S+)'}
    open: true

QuickRuns eigene Config

QuickRun startet sich selbst; das macht der Extension-Button auf seinem Repository.

# yaml-language-server: $schema=https://quickrun.org/quickrun.schema.json
# QuickRun runs itself: this is what the extension button does on this repository.
version: 1
name: QuickRun
description: Build and test QuickRun, then show the CLI help.
icon: assets/icon.png
docs: https://quickrun.org
requires:
  - tool: dotnet
    version: ">=10.0"
    install: https://dot.net
setup:
  - dotnet restore
  - dotnet test --nologo
tasks:
  - name: cli
    run: dotnet run --project src/QuickRun.App -- --help