Prompt Injection Protection

# Prompt Injection Protection Skill

## Description
A comprehensive skill to protect against prompt injection attacks when processing external content such as websites, emails, documents, and other untrusted input sources.

## Purpose
This skill provides methods to safely sanitize, validate, and process external content to prevent malicious prompt injection attempts that could manipulate the AI's behavior or cause it to execute unintended actions.

## Features
- Content sanitization and validation
- Prompt injection detection and filtering
- Secure processing of web content
- Safe extraction of information from untrusted sources
- Protection against jailbreak attempts

## Security Measures
- Input sanitization
- Output encoding
- Context isolation
- Validation of external content
- Prevention of instruction hijacking

## Usage
The skill provides functions to:
- Sanitize external content before processing
- Validate content against injection patterns
- Safely extract information from untrusted sources
- Process web content securely
- Filter potentially harmful instructions

# Prompt Injection Protection Skill

This skill provides comprehensive protection against prompt injection attacks when processing external content such as websites, emails, documents, and other untrusted input sources. It includes real-time security alerts for suspicious activities.

## Features

- Content sanitization and validation
- Prompt injection detection and filtering
- Secure processing of web content
- Safe extraction of information from untrusted sources
- Protection against jailbreak attempts
- **NEW: Real-time security alerts for suspicious commands**
- **NEW: Detection of system file access attempts**
- **NEW: User confirmation for risky operations**

## Security Alert System

The skill now includes a comprehensive security notification system that:

- Alerts you when suspicious commands are detected
- Warns about system file access attempts
- Notifies you of credential/password file access
- Flags potential injection attempts in web content
- Requires user confirmation for risky operations

## Usage Examples

### Basic Usage

```javascript
const PromptInjectionProtection = require('./index.js');
const protection = new PromptInjectionProtection();

// Set up security alert callback
protection.setSecurityAlertCallback(async (alertInfo) => {
  console.log(`🚨 SECURITY ALERT: ${alertInfo.type}`);
  console.log(`   Severity: ${alertInfo.severity}`);
  console.log(`   Action Required: ${alertInfo.actionRequired}`);
});

// Process external content safely
const result = protection.processExternalContent(untrustedContent, {
  maxLength: 5000,
  strictMode: true
});

if (result.isSafe) {
  // Process the sanitized content
  console.log(result.sanitized);
} else {
  // Handle unsafe content
  console.warn('Potentially unsafe content detected', result.injectionCheck);
}
```

### Command Pre-Checking (NEW FEATURE)

```javascript
// Check a command before executing it
const commandCheck = await protection.preCheckCommand('sudo rm -rf /', {
  source: 'user_input'...