Official Node.js SDK for the SVGMaker API, providing a clean, type-safe interface for generating, editing, and converting SVG graphics using AI.

v1.0.0 includes breaking changes. If you're upgrading from v0.x, see the Migration Guide for full details. Top breaking changes:

• Base URL changed — now targets api.svgmaker.io (automatic if using defaults)
• client.convert restructured — old client.convert.configure(...) is now client.convert.aiVectorize.configure(...), plus new sub-clients for tracing, format conversion, and batch operations
• styleParams redesigned — new style values (flat, line_art, engraving, ghibli, ...), advanced block removed, new text field added
• Aspect ratio wide/tall removed — only auto, portrait, landscape, square remain
• model parameter: New option on generate and edit — mutually exclusive with quality
• Cloud storage no longer default — generate, edit, and AI vectorize no longer store results to cloud by default; pass storage: true to persist

• Complete API Coverage: Generate, edit, convert, optimize, and manage SVGs
• TypeScript Native: Full type safety with comprehensive type definitions
• Automatic Retries: Built-in retry logic with exponential backoff
• Streaming Support: Real-time progress updates via Server-Sent Events
• Input Validation: Zod-based schema validation for all requests
• Dual Package: ESM and CommonJS support

npm install @genwave/svgmaker-sdk

Requires Node.js 18.0.0 or higher.

import { SVGMakerClient } from '@genwave/svgmaker-sdk';

const client = new SVGMakerClient('your-api-key');

// Generate an SVG
const result = await client.generate
  .configure({
    prompt: 'A minimalist mountain landscape',
    quality: 'high',
    svgText: true,
  })
  .execute();

console.log('SVG URL:', result.svgUrl);
console.log('Credits used:', result.creditCost);

// Edit an existing image
const edited = await client.edit
  .configure({
    image: './input.png',
    prompt: 'Add a red border',
    quality: 'medium',
  })
  .execute();

// AI-powered vectorization
const vectorized = await client.convert.aiVectorize
  .configure({ file: './photo.jpg', svgText: true })
  .execute();

// Stream progress for long operations
const stream = client.generate
  .configure({ prompt: 'A detailed cityscape', quality: 'high' })
  .stream();

for await (const event of stream) {
  if (event.status === 'complete') console.log('Done:', event.svgUrl);
}

Most clients use the configure().execute() pattern. generations, gallery, and account use direct methods (e.g., client.generations.list(), client.account.getInfo()).

import { SVGMakerClient, Errors } from '@genwave/svgmaker-sdk';

try {
  const result = await client.generate
    .configure({ prompt: 'A landscape' })
    .execute();
} catch (error) {
  if (error instanceof Errors.RateLimitError) {
    console.error('Rate limited. Retry after:', error.retryAfter, 'seconds');
  } else if (error instanceof Errors.InsufficientCreditsError) {
    console.error('Insufficient credits. Required:', error.creditsRequired);
  } else if (error instanceof Errors.AuthError) {
    console.error('Authentication failed:', error.message);
  } else if (error instanceof Errors.APIError) {
    console.error('API error:', error.message, error.statusCode);
  }
}

See the SDK Documentation for the full list of error types.

• Migration Guide - Upgrading from v0.x to v1.0.0
• SDK Documentation - Full SDK reference with detailed examples for every client
• API Documentation - REST API reference with endpoint details and credit costs
• Legacy API Documentation - Pre-v1 API reference

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

MIT License - see the LICENSE file for details.

• Issues: GitHub Issues
• Repository: GitHub Repository