Documentation

ziew init

Create a new Ziew project.

ziew init <project-name> [options]

Options:
  --style=<name>    Include a CSS framework (pico, water, simple, mvp, tailwind)
  --template=<name> Use a game template (phaser, kaplay, three)

Examples:
  ziew init myapp
  ziew init myapp --style=pico
  ziew init myapp --style=tailwind
  ziew init mygame --template=phaser

ziew dev

Start development server with hot reload.

ziew dev

Watches for changes in HTML, CSS, and JS files and
automatically reloads the webview.

ziew build

Build the project for the current platform.

ziew build [options]

Options:
  --release    Build with optimizations (smaller binary)

Examples:
  ziew build
  ziew build --release

ziew ship

Build optimized binaries for all platforms (cross-compilation). Always uses release optimizations.

ziew ship [options]

Options:
  --target=<target>    Build for specific target only

Targets:
  windows, macos, macos-x64, macos-arm64, linux, all

Note: ship always builds with ReleaseSmall optimization.
Use 'ziew build --release' for single-platform release builds.

Examples:
  ziew ship                   # All platforms
  ziew ship --target=windows  # Windows only
  ziew ship --target=macos    # macOS x64 only

ziew plugin

Manage project plugins. Plugins are configured in ziew.zon.

ziew plugin <command> [plugins...]

Commands:
  add <plugins...>     Enable plugins for this project
  remove <plugins...>  Disable plugins
  list                 Show enabled and available plugins

Examples:
  ziew plugin list
  ziew plugin add sqlite notify
  ziew plugin remove notify

# Create project with Pico CSS
ziew init myapp --style=pico

# Or with Tailwind
ziew init myapp --style=tailwind

// Read a file
const content = await ziew.fs.readFile('./data.txt');

// Write a file
await ziew.fs.writeFile('./output.txt', content);

// List directory
const files = await ziew.fs.readDir('./docs');

// Check if AI is available
if (ziew.ai.available()) {
  console.log('AI ready!');
}

// Text completion (returns full response)
const response = await ziew.ai.complete('Explain quantum computing', {
  maxTokens: 256  // Optional, default: 256
});

// Streaming response (async generator)
for await (const token of ziew.ai.stream(prompt, { maxTokens: 512 })) {
  output.textContent += token;
}

# Install llama.cpp
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp
cmake -B build -DLLAMA_CURL=OFF
cmake --build build
sudo cmake --install build

# Enable AI plugin
ziew plugin add ai

# Create models directory
mkdir -p ~/.ziew/models

# Download a model (example: TinyLlama)
wget -O ~/.ziew/models/tinyllama.gguf \
  https://huggingface.co/TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF/resolve/main/tinyllama-1.1b-chat-v1.0.Q4_K_M.gguf

// Check if Lua is available
if (ziew.lua.available()) {
  console.log('Lua ready!');
}

// Call a Lua function (returns string result)
const result = await ziew.lua.call('processData', arg1, arg2);

-- backend.lua
function greet(name)
  return "Hello, " .. tostring(name) .. "!"
end

function processData(a, b)
  return tostring(tonumber(a) + tonumber(b))
end

# Ubuntu/Debian
sudo apt install libluajit-5.1-dev

# Enable Lua plugin
ziew plugin add lua

// Run a command
const output = await ziew.shell.exec('git status');

// Spawn a process
const proc = await ziew.shell.spawn('npm', ['install']);

// Open file picker
const path = await ziew.dialog.open({ filters: ['*.txt', '*.md'] });

// Save dialog
const savePath = await ziew.dialog.save({ defaultName: 'document.txt' });

// Alert
await ziew.dialog.alert('Operation complete!');

console.log(ziew.platform); // 'linux', 'macos', or 'windows'
console.log(ziew.version);  // '0.2.0'