CleanFlux-AI JavaScript SDK

Official JavaScript SDK for the CleanFlux-AI text cleaning and content moderation API.

Features

• 🧹 Text Cleaning - Remove unwanted content, normalize formatting
• 🔗 URL Extraction - Extract and validate URLs from text
• 🤬 Profanity Filter - Remove or mask inappropriate language
• 📊 Text Metadata - Word count, reading time, and more
• 📝 Normalization - Whitespace, unicode, and case normalization

Installation

npm install cleanflux-ai

Requirements

• Node.js 18 or higher (uses native fetch)

Quick Start

import CleanFlux from 'cleanflux-ai';

// Initialize with your API key
const client = new CleanFlux('sk_live_your_api_key');

// Clean some text
const result = await client.clean({
  text: 'Check out   http://spam.com  for FREE stuff!!!',
  options: {
    removeUrls: true,
    normalizeWhitespace: true
  }
});

console.log(result.cleaned);
// Output: 'Check out for FREE stuff!'

API Reference

Constructor

const client = new CleanFlux(apiKey, options);

| Parameter | Type | Required | Description | |-----------|------|----------|-------------| | apiKey | string | Yes | Your CleanFlux API key | | options.baseUrl | string | No | Custom API base URL (default: https://api.cleanflux.ai) |

Methods

clean(payload)

Clean text by removing unwanted content and normalizing formatting.

const result = await client.clean({
  text: 'Your text here',
  options: {
    removeUrls: true,
    normalizeWhitespace: true,
    // ... more options
  }
});

normalize(payload)

Normalize text formatting (whitespace, unicode, case, etc.).

const result = await client.normalize({
  text: '  Hello   World  ',
  options: {
    trim: true,
    collapseWhitespace: true
  }
});
// result.normalized = 'Hello World'

extractUrls(payload)

Extract URLs from text.

const result = await client.extractUrls({
  text: 'Visit https://cleanflux.ai and https://github.com'
});
// result.urls = ['https://cleanflux.ai', 'https://github.com']

removeProfanity(payload)

Remove or mask profanity from text.

const result = await client.removeProfanity({
  text: 'Some text with bad words',
  options: {
    replacement: '***'
  }
});

metadata(payload)

Extract metadata from text.

const result = await client.metadata({
  text: 'This is a sample paragraph with several words.'
});
// result.wordCount = 8
// result.readingTime = '1 min'

ping()

Check API connectivity.

const result = await client.ping();
// result.status = 'ok'

Error Handling

The SDK throws descriptive errors for failed requests:

try {
  const result = await client.clean({ text: 'Hello' });
} catch (error) {
  console.error(error.message);   // Human-readable error message
  console.error(error.status);    // HTTP status code
  console.error(error.response);  // Full API response
}

Authentication

Get your API key from the CleanFlux Dashboard.

API keys start with sk_live_ for production use.

Documentation

Full API documentation is available at https://cleanflux.ai/docs.

Support

• 📧 Email: [email protected]
• 🐛 Issues: GitHub Issues

License

MIT License - see LICENSE for details.

Made with ❤️ by Kyros Groupe Ltd