v3.3.0

 ██████╗ ██╗     ██╗███╗   ██╗
██╔════╝ ██║     ██║████╗  ██║
██║  ███╗██║     ██║██╔██╗ ██║
██║   ██║██║     ██║██║╚██╗██║
╚██████╔╝███████╗██║██║ ╚████║
 ╚═════╝ ╚══════╝╚═╝╚═╝  ╚═══╝

██████╗ ██████╗  ██████╗ ███████╗ █████╗ ███╗   ██╗██╗████████╗██╗   ██╗
██╔══██╗██╔══██╗██╔═══██╗██╔════╝██╔══██╗████╗  ██║██║╚══██╔══╝╚██╗ ██╔╝
██████╔╝██████╔╝██║   ██║█████╗  ███████║██╔██╗ ██║██║   ██║    ╚████╔╝
██╔═══╝ ██╔══██╗██║   ██║██╔══╝  ██╔══██║██║╚██╗██║██║   ██║     ╚██╔╝
██║     ██║  ██║╚██████╔╝██║     ██║  ██║██║ ╚████║██║   ██║      ██║
╚═╝     ╚═╝  ╚═╝ ╚═════╝ ╚═╝     ╚═╝  ╚═╝╚═╝  ╚═══╝╚═╝   ╚═╝      ╚═╝

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
                    v3.3.0 · AI Ecosystem Release
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

---

Why Glin Profanity Over AI-Native Moderation?

"Why not just let GPT/Claude/Gemini handle profanity detection?"

┌─────────────────────────────────────────────────────────────────────────┐
│                    THE SMART MODERATION APPROACH                        │
├─────────────────────────────────────────────────────────────────────────┤
│                                                                         │
│   User Message                                                          │
│        │                                                                │
│        ▼                                                                │
│   ┌─────────────────┐                                                   │
│   │  Glin Profanity │  ◄─── Fast, Free, Local (<1ms)                    │
│   │   First Pass    │                                                   │
│   └────────┬────────┘                                                   │
│            │                                                            │
│      ┌─────┴─────┐                                                      │
│      │           │                                                      │
│      ▼           ▼                                                      │
│   ┌──────┐   ┌──────────┐                                               │
│   │ SAFE │   │ FLAGGED  │                                               │
│   └──┬───┘   └────┬─────┘                                               │
│      │            │                                                     │
│      │      ┌─────┴─────┐                                               │
│      │      │           │                                               │
│      │      ▼           ▼                                               │
│      │   ┌──────┐   ┌────────────┐                                      │
│      │   │BLOCK │   │  UNCERTAIN │                                      │
│      │   └──────┘   └─────┬──────┘                                      │
│      │                    │                                             │
│      │                    ▼                                             │
│      │            ┌───────────────┐                                     │
│      │            │   AI Review   │  ◄─── Only when needed              │
│      │            │ (GPT/Claude)  │       (saves 90%+ API costs)        │
│      │            └───────┬───────┘                                     │
│      │                    │                                             │
│      ▼                    ▼                                             │
│   ┌─────────────────────────────┐                                       │
│   │      FINAL DECISION         │                                       │
│   └─────────────────────────────┘                                       │
│                                                                         │
└─────────────────────────────────────────────────────────────────────────┘

The Numbers

Metric	AI-Only (GPT/Claude)	Glin Profanity	Hybrid Approach
Cost per 1K checks	$10-30	$0	~$1-3
Latency	200-500ms	<1ms	<5ms avg
Rate Limits	Yes	No	Minimal
Works Offline	No	Yes	Partial
Consistency	Variable	Deterministic	Deterministic

---

🆕 New Packages

🤖 glin-profanity-mcp

MCP Server for AI Assistants — Claude Desktop, Cursor, Windsurf, and more.

npx glin-profanity-mcp          # STDIO mode (local)
npx glin-profanity-mcp-http     # HTTP mode (cloud)

Capabilities: 19 Tools · 4 Prompts · 5 Resources

📖 MCP Documentation

---

🦞 openclaw-profanity

OpenClaw/Moltbot Plugin — Content moderation for AI agents on messaging platforms.

npm install openclaw-profanity

📖 OpenClaw Guide

---

🧠 AI & ML Features

ML Toxicity Detection

TensorFlow.js-powered toxicity detection beyond keywords.

import { loadToxicityModel, checkToxicity } from 'glin-profanity/ml';

await loadToxicityModel({ threshold: 0.9 });
const result = await checkToxicity("You're the worst player ever");
// Detects: toxicity, insults, threats, identity attacks, obscene content

Semantic Analysis

Embeddings-based detection for content without obvious profanity.

import { createSemanticAnalyzer } from 'glin-profanity/ai/semantic';

const analyzer = createSemanticAnalyzer({ 
  embeddingProvider: openaiProvider,
  threshold: 0.7 
});

const result = await analyzer.analyze('You should uninstall life');
// Catches veiled threats, sarcastic insults, context-dependent toxicity

Hybrid Detection

Combine keyword + ML + semantic for 99.8%+ accuracy.

import { HybridFilter } from 'glin-profanity/ml';

const hybrid = new HybridFilter({
  enableKeyword: true,
  enableML: true,
  enableSemantic: true
});

📖 AI & ML Features

---

🔌 AI Framework Integrations

┌────────────────────────────────────────────────────────────────────┐
│                    SUPPORTED AI FRAMEWORKS                         │
├────────────────────────────────────────────────────────────────────┤
│                                                                    │
│   ┌─────────────┐  ┌─────────────┐  ┌─────────────┐               │
│   │   OpenAI    │  │  LangChain  │  │  Vercel AI  │               │
│   │  gpt-4o     │  │   Agents    │  │    SDK      │               │
│   └─────────────┘  └─────────────┘  └─────────────┘               │
│                                                                    │
│   ┌─────────────┐  ┌─────────────┐  ┌─────────────┐               │
│   │  OpenClaw   │  │    MCP      │  │  Semantic   │               │
│   │  Moltbot    │  │   Server    │  │  Kernel     │               │
│   └─────────────┘  └─────────────┘  └─────────────┘               │
│                                                                    │
└────────────────────────────────────────────────────────────────────┘

OpenAI Function Calling

import { profanityTools } from 'glin-profanity/ai/openai';
const response = await openai.chat.completions.create({
  model: 'gpt-4o',
  tools: profanityTools
});

LangChain Tools

import { allProfanityTools } from 'glin-profanity/ai/langchain';
const agent = createReactAgent({ llm: model, tools: allProfanityTools });

Vercel AI SDK

import { profanityTools } from 'glin-profanity/ai/vercel';
const result = await generateText({ model: openai('gpt-4o'), tools: profanityTools });

📖 All Integrations

---

⚛️ Framework Support

React Hook

import { useProfanityChecker } from 'glin-profanity';

function ChatInput() {
  const { result, checkText } = useProfanityChecker({ detectLeetspeak: true });
  return <input onChange={(e) => checkText(e.target.value)} />;
}

Supported Frameworks

• ✅ React / Next.js (App & Pages Router)
• ✅ Vue / Nuxt
• ✅ Angular / Svelte
• ✅ Express / Fastify / Hono
• ✅ Flask / Django (Python)

Serverless

• ✅ AWS Lambda
• ✅ Google Cloud Functions
• ✅ Azure Functions
• ✅ Vercel Edge Functions
• ✅ Cloudflare Workers

---

📚 Documentation

Category	Guides
Getting Started	Installation · Quick Reference · Configuration
MCP Server	Overview · Setup · Tools
Integrations	OpenAI · LangChain · Vercel AI · OpenClaw · Semantic
Advanced	Features · Performance · Troubleshooting
**Prod...

v3.2.2

🚀 Stable Release v3.2.2

📦 Installation

JavaScript/TypeScript:

npm install glin-profanity@stable

Python:

pip install glin-profanity

🔄 Changes

• docs(openclaw): enhance README with banner, SEO, and marketing content
• fix(ci): build glin-profanity dependency before MCP package

📋 Package Information

• npm tag: stable
• PyPI classifier: Development Status :: 5 - Production/Stable
• Cross-platform: Identical APIs for JavaScript and Python

🔗 Links

• npm package
• PyPI package
• Documentation

---

🤖 This release was automatically generated

v3.2.0

🚀 Stable Release v3.2.0

📦 Installation

JavaScript/TypeScript:

npm install glin-profanity@stable

Python:

pip install glin-profanity

🔄 Changes

• feat: add openclaw-profanity, glin-profanity-mcp, and multi-modal support
• fix: remove force-include causing duplicate filenames in wheel
• fix: rename Norwegian.json to lowercase for case-sensitive builds

📋 Package Information

• npm tag: stable
• PyPI classifier: Development Status :: 5 - Production/Stable
• Cross-platform: Identical APIs for JavaScript and Python

🔗 Links

• npm package
• PyPI package
• Documentation

---

🤖 This release was automatically generated

openclaw-profanity v1.0.3

🦞 OpenClaw Profanity Plugin v1.0.3

Content moderation plugin for OpenClaw (formerly Moltbot/Clawdbot).

📦 Installation

# Via npm
npm install openclaw-profanity

# Via OpenClaw CLI
openclaw plugins install openclaw-profanity

✨ Features

• 4 Integration Methods: Skills, Plugin API, Hooks, MCP
• 24 Languages: Full multilingual profanity detection
• Platform Support: WhatsApp, Telegram, Discord, Slack, iMessage
• Leetspeak Detection: Catches obfuscated profanity
• Context-Aware: Understands appropriate word usage

🔗 Links

• npm package
• Documentation
• glin-profanity core

---

🤖 This release was automatically generated

openclaw-profanity v1.0.1

🦞 OpenClaw Profanity Plugin v1.0.1

Content moderation plugin for OpenClaw (formerly Moltbot/Clawdbot).

📦 Installation

# Via npm
npm install openclaw-profanity

# Via OpenClaw CLI
openclaw plugins install openclaw-profanity

✨ Features

• 4 Integration Methods: Skills, Plugin API, Hooks, MCP
• 24 Languages: Full multilingual profanity detection
• Platform Support: WhatsApp, Telegram, Discord, Slack, iMessage
• Leetspeak Detection: Catches obfuscated profanity
• Context-Aware: Understands appropriate word usage

🔗 Links

• npm package
• Documentation
• glin-profanity core

---

🤖 This release was automatically generated

glin-profanity-mcp v1.0.3

🤖 Glin Profanity MCP Server v1.0.3

MCP (Model Context Protocol) server for AI assistants like Claude, Cursor, and Windsurf.

📦 Installation

npm install -g glin-profanity-mcp

⚡ Quick Setup

Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "glin-profanity": {
      "command": "npx",
      "args": ["-y", "glin-profanity-mcp"]
    }
  }
}

✨ Features

• 12 Tools: Check, censor, analyze, batch process profanity
• 24 Languages: Full multilingual support
• Streaming HTTP: Real-time content moderation
• Context-Aware: Domain-specific whitelists
• AI-Optimized: Designed for LLM tool use

🔗 Links

• npm package
• Documentation
• glin-profanity core

---

🤖 This release was automatically generated

glin-profanity-mcp v1.0.2

🤖 Glin Profanity MCP Server v1.0.2

MCP (Model Context Protocol) server for AI assistants like Claude, Cursor, and Windsurf.

📦 Installation

npm install -g glin-profanity-mcp

⚡ Quick Setup

Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "glin-profanity": {
      "command": "npx",
      "args": ["-y", "glin-profanity-mcp"]
    }
  }
}

✨ Features

• 12 Tools: Check, censor, analyze, batch process profanity
• 24 Languages: Full multilingual support
• Streaming HTTP: Real-time content moderation
• Context-Aware: Domain-specific whitelists
• AI-Optimized: Designed for LLM tool use

🔗 Links

• npm package
• Documentation
• glin-profanity core

---

🤖 This release was automatically generated

glin-profanity-mcp v1.0.1

🤖 Glin Profanity MCP Server v1.0.1

MCP (Model Context Protocol) server for AI assistants like Claude, Cursor, and Windsurf.

📦 Installation

npm install -g glin-profanity-mcp

⚡ Quick Setup

Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "glin-profanity": {
      "command": "npx",
      "args": ["-y", "glin-profanity-mcp"]
    }
  }
}

✨ Features

• 12 Tools: Check, censor, analyze, batch process profanity
• 24 Languages: Full multilingual support
• Streaming HTTP: Real-time content moderation
• Context-Aware: Domain-specific whitelists
• AI-Optimized: Designed for LLM tool use

🔗 Links

• npm package
• Documentation
• glin-profanity core

---

🤖 This release was automatically generated

v3.1.5

🚀 Stable Release v3.1.5

📦 Installation

JavaScript/TypeScript:

npm install glin-profanity@stable

Python:

pip install glin-profanity

🔄 Changes

• fix: dynamic versioning and update tool count to 19
• feat: add openclaw-profanity, glin-profanity-mcp, and multi-modal support
• fix: remove force-include causing duplicate filenames in wheel
• fix: rename Norwegian.json to lowercase for case-sensitive builds

📋 Package Information

• npm tag: stable
• PyPI classifier: Development Status :: 5 - Production/Stable
• Cross-platform: Identical APIs for JavaScript and Python

🔗 Links

• npm package
• PyPI package
• Documentation

---

🤖 This release was automatically generated

v3.1.0

🚀 Stable Release v3.1.0

📦 Installation

JavaScript/TypeScript:

npm install glin-profanity@stable

Python:

pip install glin-profanity

🔄 Changes

• feat: v3.0.0 - ML integration, leetspeak detection, Unicode normalization
• fix: add Dutch language support and fix Turkish dictionary (#76)

📋 Package Information

• npm tag: stable
• PyPI classifier: Development Status :: 5 - Production/Stable
• Cross-platform: Identical APIs for JavaScript and Python

🔗 Links

• npm package
• PyPI package
• Documentation

---

🤖 This release was automatically generated