🛡️ profanity-nsfw-violence-checker

A robust and customizable TypeScript/JavaScript library for detecting and censoring profanity, sexual content, violence, and hate speech in plain text and HTML.

✨ Features

• 🚨 Detects profanity, sexual content, violence, and hate speech
• 🧠 Supports leetspeak & misspellings (e.g., s3x, f@ck)
• 🧪 Customizable severity levels (low/medium/high)
• 🛡️ Safe HTML censoring without breaking markup
• 📦 Built-in TypeScript types and full support
• 🧰 Add/remove words, enable whitelist, toggle strictMode

📦 Installation

npm install profanity-nsfw-violence-checker

yarn add profanity-nsfw-violence-checker

🚀 Quick Start

import { ProfanityChecker } from 'profanity-nsfw-violence-checker';

const checker = new ProfanityChecker();

const result = checker.check("That fucking movie was damn violent!");
console.log(result);

/* Output:
{
  originalText: 'That fucking movie was a damn disgrace. It was so violent!',
  isFlagged: true,
  isProfane: true,
  isSexual: false,
  isViolent: true,
  isHateSpeech: false,
  severity: 'high',
  matches: {
    profanity: [ { word: 'fucking', position: 5, severity: 'medium', ... }, { word: 'damn', position: 28, severity: 'medium', ... } ],
    sexual: [],
    violence: [ { word: 'violent', position: 51, severity: 'high', ... } ],
    hateSpeech: []
  }
}
*/

const censored = checker.censor("That fucking movie was damn violent!");
console.log(censored); // Output: That ******* movie was **** violent!

🖼️ HTML Censoring

const html = '<p>This is <strong>damn</strong> violent.</p>';
const cleanHtml = checker.censorHtml(html);
console.log(cleanHtml);
// <p>This is <strong>****</strong> violent.</p>

⚙️ Configuration

const customChecker = new ProfanityChecker({
  censorCharacter: '#',
  strictMode: true,
  allowWhitelist: true,
  customWords: {
    profanity: ['frick', 'dang'],
    hateSpeech: ['subhuman']
  }
});

📚 API Methods

• checker.check(text: string): CheckResult → Analyze text and get details
• checker.censor(text: string): string → Censor words in plain text
• checker.censorHtml(html: string): string → Censor inside HTML safely
• checker.addWords(category, words, severity?) → Add words to a category
• checker.removeWords(category, words) → Remove words from category
• checker.addToWhitelist(words) → Exclude words from detection
• checker.getStats() → Dictionary stats (count per category)

🧠 Supported Categories

• profanity: e.g., f***, s***, etc.
• sexual: e.g., p***, blowjob, etc.
• violence: e.g., kill, stab, etc.
• hateSpeech: e.g., racial slurs, subhuman, etc.

🧪 TypeScript Support

type Severity = 'low' | 'medium' | 'high';

interface Match {
  word: string;
  position: number;
  severity: Severity;
  context: string;
}

interface CheckResult {
  originalText: string;
  isFlagged: boolean;
  isProfane: boolean;
  isSexual: boolean;
  isViolent: boolean;
  isHateSpeech: boolean;
  severity: Severity;
  matches: {
    profanity: Match[];
    sexual: Match[];
    violence: Match[];
    hateSpeech: Match[];
  };
}

📈 Example Output

{
  isFlagged: true,
  isProfane: true,
  isSexual: false,
  isViolent: true,
  isHateSpeech: false,
  severity: 'high',
  matches: {
    profanity: [{ word: 'fucking', position: 5, severity: 'high' }],
    sexual: [],
    violence: [{ word: 'violent', position: 42, severity: 'high' }],
    hateSpeech: []
  }
}

🙌 Contributions

PRs and issues are welcome. Help improve detection accuracy, word lists, or features!

📄 License

MIT License – Victor Olayemi