openclaw-profanity

What This Skill Does

The openclaw-profanity skill acts as a robust content moderation layer for OpenClaw and Moltbot agents. Designed to maintain community standards, it intercepts incoming messages to identify and handle profanity in real-time. The plugin is highly sophisticated, utilizing detection algorithms that account for obfuscation techniques like leetspeak (e.g., converting 'f4ck' to 'fuck') and Unicode lookalikes (e.g., identifying Cyrillic characters used to bypass simple text filters). It supports multiple languages and offers flexible enforcement actions including warning users, censoring specific words with custom characters, blocking entire messages to prevent toxicity, or silently logging violations for moderation audit trails.

Installation

To integrate this skill into your agent, use the OpenClaw CLI tool: clawhub install openclaw/skills/skills/thegdsks/openclaw-profanity

Alternatively, install via npm for manual integration: npm install openclaw-profanity

Use Cases

This skill is essential for developers building safe digital environments. Common use cases include:

• Discord community management: Automatically filtering slurs or toxic language from high-traffic chat servers.
• Corporate Helpdesks: Preventing abusive language from reaching live support agents on Slack or Telegram.
• Educational Bots: Ensuring a clean, family-friendly interaction space for younger users.
• Moderation Dashboards: Using the 'log' and 'custom' actions to build automated ban-tracking systems for repeat offenders.

Example Prompts

1. "OpenClaw, enable strict profanity filtering and block any messages containing offensive language immediately."
2. "Update my agent configuration to censor all profanity with asterisks and ignore my custom list of safe-but-flagged words."
3. "Set up a custom violation handler that bans a user on their fourth attempt to use restricted language in the Discord channel."

Tips & Limitations

• Performance: While the plugin is optimized, enabling Unicode normalization on extremely high-velocity chat streams may slightly increase latency.
• Whitelisting: Use the ignoreWords array to prevent "false positives" if your community uses specific slang or reclaimed words that the system might otherwise flag.
• Hierarchy: Always test your custom violation logic in a staging environment before deploying to production, especially when configuring automated user bans or administrative actions to avoid accidental account lockouts.