/**
* Profanity Guard Hook for OpenClaw
*
* This hook automatically moderates all messages passing through
* OpenClaw, including WhatsApp, Telegram, Discord, Slack, iMessage, etc.
*
* @remarks
* Compatible with OpenClaw, Moltbot (legacy), and Clawdbot (legacy).
*
* Installation:
* ```bash
* openclaw hooks install openclaw-profanity
* ```
*
* Or manually add to your hooks directory:
* ```bash
* cp -r node_modules/openclaw-profanity/hooks/profanity-guard ~/.openclaw/hooks/
* ```
*/
import { createProfanityGuard } from '../../src/hooks';
/**
* Default hook configuration
*
* Customize by editing this file or providing environment variables:
* - PROFANITY_MODE: 'strict' | 'moderate' | 'lenient'
* - PROFANITY_BLOCK: 'true' | 'false'
* - PROFANITY_CENSOR: 'true' | 'false'
* - PROFANITY_LANGUAGES: 'english,spanish,french'
*/
const config = {
mode: (process.env.PROFANITY_MODE as 'strict' | 'moderate' | 'lenient') || 'moderate',
blockProfanity: process.env.PROFANITY_BLOCK === 'true',
censorProfanity: process.env.PROFANITY_CENSOR !== 'false', // Default true
languages: process.env.PROFANITY_LANGUAGES?.split(',') || ['english'],
logIncidents: true,
};
// Create the guard
const guard = createProfanityGuard(config);
/**
* Hook handler - called by OpenClaw for each message
*/
export async function handler(message: string, context?: Record<string, unknown>) {
const result = await guard.processMessage({
message,
channel: context?.channel as string,
userId: context?.userId as string,
direction: context?.direction as 'incoming' | 'outgoing',
});
return result;
}
/**
* Hook metadata
*/
export const metadata = {
name: 'profanity-guard',
description: 'Automatic content moderation with 24-language profanity detection',
version: '1.0.0',
author: 'GLINR',
config: {
mode: {
type: 'string',
enum: ['strict', 'moderate', 'lenient'],
default: 'moderate',
description: 'Moderation strictness level',
},
blockProfanity: {
type: 'boolean',
default: false,
description: 'Block messages containing profanity',
},
censorProfanity: {
type: 'boolean',
default: true,
description: 'Automatically censor profanity',
},
languages: {
type: 'array',
default: ['english'],
description: 'Languages to check for profanity',
},
},
};
export default handler;
