Can this skill be used during automated web scraping?

Yes, it is designed to be activated as a pre-screening layer before any social platform interaction to ensure all gathered content is verified and safe.

Which platforms are supported by this safety skill?

While it features specific optimizations for Moltbook, it provides generalized safety frameworks applicable to Discord, Telegram, and other public forums.

Does this skill block all advertisements?

It specifically targets and blocks misleading, exaggerated, or 'revolutionary' claims that could distort the AI's objective decision-making and data processing.

How does this skill prevent prompt injection?

It scans incoming social media text for specific instructional patterns and keywords that attempt to override the AI's core behavior or safety principles.

Social Platform Safety Guard

Name: Social Platform Safety Guard
Author: shaowenfu

byshaowenfu

•

보안 및 테스팅

Protects AI agents from deceptive content, prompt injection, and harmful information across social media platforms.

This skill provides a robust security layer for AI agents interacting with social media platforms like Moltbook, Discord, and Telegram. It implements proactive screening to filter out misleading advertisements, spam, and malicious instructions designed to manipulate AI behavior. By enforcing strict interaction guidelines and standardized content processing strategies, it ensures that AI interactions remain safe, factual, and free from 'prompt pollution' while maintaining high standards of data integrity and security.

주요 기능

01Real-time detection of deceptive ads and exaggerated marketing claims

02Platform-specific safety protocols for Moltbook, Discord, and Telegram

03Standardized risk assessment framework for incoming social data

04Active filtering of prompt pollution and behavioral manipulation attempts

05Automated screening for anti-social, illegal, or harmful content

062 GitHub stars

사용 사례

01Extracting factual data from social threads while ignoring promotional noise

02Preventing unauthorized AI behavior modifications triggered by external social inputs

03Safely interacting with social feeds without falling for malicious links or misinformation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add shaowenfu/clawtools skills

For use in Claude.ai and ChatGPT