A simple, fast, and extensible JavaScript package to detect and censor abusive words in multiple Indian and international languages. Useful for chat moderation, content filtering, and building safe online communities.

• Detects and censors abusive words in Hindi, English, Bengali, Urdu, and more
• Customizable censorship character (grawlix)
• Fine-grained control with alwaysAllow and alwaysBlock word lists
• Easy to use and integrate in Node.js projects

npm install cleanword

const { cleanText } = require('cleanword');

const options = {
  language: ['english', 'hindi'],
  grawlixChar: '@',
  alwaysAllow: ['kutto'],
  alwaysBlock: ['test', 'what'],
};
const cleaned = cleanText('This is a test sentence with kutto and what.', options);
console.log(cleaned); // This is a @@@@ sentence with kutto and @@@@..

import { cleanText } from 'cleanword';

interface CleanTextOptions {
  language: string[];
  grawlixChar: string;
  alwaysAllow: string[];
  alwaysBlock: string[];
}

const options: CleanTextOptions = {
  language: ['english', 'hindi'],
  grawlixChar: '@',
  alwaysAllow: ['kutto'],
  alwaysBlock: ['test', 'what'],
};

const cleaned: string = cleanText('This is a test sentence with kutto and what.', options);
console.log(cleaned); // This is a @@@@ sentence with kutto and @@@@. 

• text (string): The input string to clean.
• options (object, optional):
  • language: string | string[] — Language(s) to check (default: 'hindi').
  • grawlixChar: string — Character to use for censorship (default: '*').
  • alwaysAllow: string[] — Words that should never be censored, even if abusive.
  • alwaysBlock: string[] — Words that should always be censored, even if not abusive.
  • customAbuseSet: Set<string> — Custom set of abusive words (for advanced use/testing).

Returns: The cleaned string with abusive words replaced by the grawlix character.

• Hindi
• English
• Assamese
• Bengali
• Bhojpuri
• Marathi
• Chhattisgarhi
• Gujarati
• Haryanvi
• Kannada
• Kashmiri
• Konkani
• Ladakhi
• Malayalam
• Manipuri
• Marwari
• Nepali
• Odia
• Punjabi
• Rajasthani
• Tamil
• Telugu
• Urdu

You can specify one or more languages using the language option. Example:

cleanText('some text', { language: ['hindi', 'english'] });

1. Fork this repository and clone your fork.
2. Install dependencies:

   npm install

3. Add or improve abusive word lists in src/abuse_words.js.
4. Add or update tests in Test/cleanText.test.js.
5. Run tests:

   npm test

6. Submit a pull request with a clear description of your changes.

Guidelines:

• Please be respectful and avoid adding non-abusive or irrelevant words.
• Keep word lists accurate and up-to-date for each language.
• Add tests for any new features or language support.

Developed with ❤️ by Nabarup

If you find this package useful, ⭐ star the repo and share it!

MIT © 2025 Nabarup.
Use freely. Contribute with respect.

For feature requests, feedback, or bug reports, open an issue or email me at [email protected] .