Create app.py

app.py

@@ -0,0 +1,339 @@
+"""
+TSZ (Thyris Safe Zone) - Interactive Demo
+==========================================
+This demo showcases how TSZ detects and redacts sensitive information
+before it reaches LLMs or third-party APIs.
+
+GitHub: https://github.com/thyrisAI/safe-zone
+Website: https://thyris.ai
+"""
+
+import gradio as gr
+import re
+from dataclasses import dataclass
+from typing import List, Tuple
+
+# ============================================================================
+# PII Detection Patterns (Simplified version of TSZ's detection engine)
+# ============================================================================
+
+@dataclass
+class Detection:
+    """Represents a detected PII entity"""
+    type: str
+    value: str
+    start: int
+    end: int
+    confidence: str
+
+class TSZDetector:
+    """
+    Simplified PII detector that mimics TSZ's detection capabilities.
+    The real TSZ uses more sophisticated pattern matching and validators.
+    """
+    
+    PATTERNS = {
+        "EMAIL": {
+            "pattern": r'\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Z|a-z]{2,}\b',
+            "placeholder": "[EMAIL]",
+            "description": "Email addresses"
+        },
+        "PHONE": {
+            "pattern": r'\b(?:\+?1[-.\s]?)?\(?[0-9]{3}\)?[-.\s]?[0-9]{3}[-.\s]?[0-9]{4}\b',
+            "placeholder": "[PHONE]",
+            "description": "Phone numbers (US format)"
+        },
+        "PHONE_TR": {
+            "pattern": r'\b(?:\+90|0)?[-.\s]?[0-9]{3}[-.\s]?[0-9]{3}[-.\s]?[0-9]{2}[-.\s]?[0-9]{2}\b',
+            "placeholder": "[PHONE]",
+            "description": "Phone numbers (Turkey format)"
+        },
+        "CREDIT_CARD": {
+            "pattern": r'\b(?:4[0-9]{12}(?:[0-9]{3})?|5[1-5][0-9]{14}|3[47][0-9]{13}|6(?:011|5[0-9]{2})[0-9]{12})\b',
+            "placeholder": "[CREDIT_CARD]",
+            "description": "Credit card numbers"
+        },
+        "SSN": {
+            "pattern": r'\b[0-9]{3}-[0-9]{2}-[0-9]{4}\b',
+            "placeholder": "[SSN]",
+            "description": "Social Security Numbers (US)"
+        },
+        "TC_KIMLIK": {
+            "pattern": r'\b[1-9][0-9]{10}\b',
+            "placeholder": "[TC_KIMLIK]",
+            "description": "Turkish National ID (TC Kimlik)"
+        },
+        "IBAN": {
+            "pattern": r'\b[A-Z]{2}[0-9]{2}[A-Z0-9]{4}[0-9]{7}([A-Z0-9]?){0,16}\b',
+            "placeholder": "[IBAN]",
+            "description": "International Bank Account Numbers"
+        },
+        "IP_ADDRESS": {
+            "pattern": r'\b(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b',
+            "placeholder": "[IP_ADDRESS]",
+            "description": "IPv4 addresses"
+        },
+        "API_KEY": {
+            "pattern": r'\b(?:sk-[a-zA-Z0-9]{32,}|api[_-]?key[_-]?[=:]\s*["\']?[a-zA-Z0-9]{16,}["\']?)\b',
+            "placeholder": "[API_KEY]",
+            "description": "API keys and secrets"
+        },
+        "AWS_KEY": {
+            "pattern": r'\b(?:AKIA|ABIA|ACCA|ASIA)[A-Z0-9]{16}\b',
+            "placeholder": "[AWS_ACCESS_KEY]",
+            "description": "AWS Access Keys"
+        },
+        "PASSWORD": {
+            "pattern": r'(?:password|passwd|pwd)[_\s]*[=:]\s*["\']?[^\s"\']{4,}["\']?',
+            "placeholder": "[PASSWORD]",
+            "description": "Passwords in config/code"
+        },
+        "JWT_TOKEN": {
+            "pattern": r'\beyJ[A-Za-z0-9-_]+\.eyJ[A-Za-z0-9-_]+\.[A-Za-z0-9-_]+\b',
+            "placeholder": "[JWT_TOKEN]",
+            "description": "JSON Web Tokens"
+        },
+        "DATE_OF_BIRTH": {
+            "pattern": r'\b(?:0[1-9]|[12][0-9]|3[01])[/-](?:0[1-9]|1[0-2])[/-](?:19|20)[0-9]{2}\b',
+            "placeholder": "[DOB]",
+            "description": "Dates of birth"
+        }
+    }
+    
+    def detect(self, text: str) -> Tuple[List[Detection], str]:
+        """
+        Detect PII in text and return detections + redacted text.
+        
+        Returns:
+            Tuple of (list of detections, redacted text)
+        """
+        detections = []
+        redacted_text = text
+        
+        for pii_type, config in self.PATTERNS.items():
+            pattern = config["pattern"]
+            placeholder = config["placeholder"]
+            
+            for match in re.finditer(pattern, text, re.IGNORECASE):
+                # Avoid duplicate detections for overlapping patterns
+                value = match.group()
+                
+                # Skip if this looks like a false positive
+                if pii_type == "TC_KIMLIK" and len(value) != 11:
+                    continue
+                    
+                detections.append(Detection(
+                    type=pii_type,
+                    value=value,
+                    start=match.start(),
+                    end=match.end(),
+                    confidence="HIGH" if len(value) > 8 else "MEDIUM"
+                ))
+        
+        # Sort by position (reverse) for correct replacement
+        detections.sort(key=lambda x: x.start, reverse=True)
+        
+        # Redact in reverse order to maintain positions
+        for det in detections:
+            placeholder = self.PATTERNS[det.type]["placeholder"]
+            redacted_text = redacted_text[:det.start] + placeholder + redacted_text[det.end:]
+        
+        # Re-sort for display (by position, ascending)
+        detections.sort(key=lambda x: x.start)
+        
+        return detections, redacted_text
+
+# ============================================================================
+# Gradio Interface
+# ============================================================================
+
+detector = TSZDetector()
+
+def analyze_text(text: str) -> Tuple[str, str, str]:
+    """
+    Main analysis function for Gradio interface.
+    
+    Returns:
+        Tuple of (redacted_text, detection_report, stats)
+    """
+    if not text.strip():
+        return "", "⚠️ Please enter some text to analyze.", ""
+    
+    detections, redacted_text = detector.detect(text)
+    
+    if not detections:
+        return text, "✅ No sensitive information detected.", "**0** PII entities found"
+    
+    # Build detection report
+    report_lines = ["### 🔍 Detected Entities\n"]
+    
+    # Group by type
+    by_type = {}
+    for det in detections:
+        if det.type not in by_type:
+            by_type[det.type] = []
+        by_type[det.type].append(det)
+    
+    for pii_type, items in by_type.items():
+        desc = detector.PATTERNS[pii_type]["description"]
+        report_lines.append(f"\n**{pii_type}** ({desc})")
+        for item in items:
+            masked_value = item.value[:3] + "***" + item.value[-2:] if len(item.value) > 5 else "***"
+            report_lines.append(f"- `{masked_value}` (confidence: {item.confidence})")
+    
+    report = "\n".join(report_lines)
+    
+    # Stats
+    stats = f"**{len(detections)}** PII entities found across **{len(by_type)}** categories"
+    
+    return redacted_text, report, stats
+
+# Example texts for users to try
+EXAMPLES = [
+    ["""Hi, my name is John Smith and my email is john.smith@company.com. 
+You can reach me at +1-555-123-4567. My SSN is 123-45-6789."""],
+    
+    ["""Customer Order #12345
+Name: Jane Doe
+Email: jane.doe@example.org
+Phone: (555) 987-6543
+Credit Card: 4532015112830366
+Shipping Address: 123 Main St, New York, NY 10001"""],
+    
+    ["""API Configuration:
+api_key = "sk-1234567890abcdefghijklmnopqrstuvwxyz"
+AWS_ACCESS_KEY: AKIAIOSFODNN7EXAMPLE
+password: "super_secret_123"
+database_url: postgres://user:pass@192.168.1.100:5432/db"""],
+    
+    ["""Merhaba, ben Ahmet Yılmaz. TC Kimlik numaram 12345678901.
+Telefon: 0532 123 45 67
+IBAN: TR330006100519786457841326
+Email: ahmet.yilmaz@sirket.com.tr"""],
+]
+
+# Custom CSS
+custom_css = """
+.gradio-container {
+    font-family: 'Inter', -apple-system, BlinkMacSystemFont, sans-serif;
+}
+.detection-box {
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    border-radius: 12px;
+    padding: 20px;
+}
+footer {
+    visibility: hidden;
+}
+"""
+
+# Build the interface
+with gr.Blocks(css=custom_css, title="TSZ - Thyris Safe Zone Demo") as demo:
+    
+    # Header
+    gr.Markdown("""
+    # 🛡️ TSZ (Thyris Safe Zone) - Interactive Demo
+    
+    **TSZ** is an open-source PII detection and guardrails engine that prevents sensitive data 
+    from leaking to LLMs and third-party APIs. This demo showcases its core detection capabilities.
+    
+    <div style="display: flex; gap: 10px; margin: 10px 0;">
+        <a href="https://github.com/thyrisAI/safe-zone" target="_blank">
+            <img src="https://img.shields.io/badge/GitHub-Repository-black?logo=github" alt="GitHub">
+        </a>
+        <a href="https://thyris.ai" target="_blank">
+            <img src="https://img.shields.io/badge/Website-thyris.ai-blue" alt="Website">
+        </a>
+        <a href="https://github.com/thyrisAI/safe-zone/blob/main/LICENSE" target="_blank">
+            <img src="https://img.shields.io/badge/License-Apache%202.0-green" alt="License">
+        </a>
+    </div>
+    """)
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            # Input
+            input_text = gr.Textbox(
+                label="📝 Input Text",
+                placeholder="Paste text containing emails, phone numbers, credit cards, API keys, passwords, etc...",
+                lines=10,
+                max_lines=20
+            )
+            
+            analyze_btn = gr.Button("🔍 Analyze & Redact", variant="primary", size="lg")
+            
+            gr.Markdown("### 📋 Try These Examples")
+            gr.Examples(
+                examples=EXAMPLES,
+                inputs=input_text,
+                label=""
+            )
+        
+        with gr.Column(scale=1):
+            # Outputs
+            stats_output = gr.Markdown(label="Statistics")
+            
+            redacted_output = gr.Textbox(
+                label="🔒 Redacted Output (Safe to send to LLMs)",
+                lines=10,
+                max_lines=20,
+                interactive=False
+            )
+            
+            detection_report = gr.Markdown(label="Detection Report")
+    
+    # Supported types info
+    gr.Markdown("""
+    ---
+    ### 🎯 Supported Detection Types
+    
+    | Category | Types |
+    |----------|-------|
+    | **Personal Info** | Email, Phone (US/TR), SSN, TC Kimlik, Date of Birth |
+    | **Financial** | Credit Cards, IBAN |
+    | **Technical** | IP Addresses, API Keys, AWS Keys, Passwords, JWT Tokens |
+    
+    ---
+    
+    ### 🚀 Get the Full Version
+    
+    This demo uses a simplified detection engine. The full **TSZ** includes:
+    
+    - ✅ More comprehensive pattern matching with validators
+    - ✅ Configurable guardrails and policies
+    - ✅ Allowlist/blocklist management
+    - ✅ REST API with hot-reloadable rules
+    - ✅ High-performance Go implementation with Redis caching
+    - ✅ Docker deployment ready
+    
+    **Install TSZ:**
+    ```bash
+    git clone https://github.com/thyrisAI/safe-zone.git
+    cd safe-zone
+    docker-compose up -d
+    ```
+    
+    ---
+    
+    <p style="text-align: center; color: #666;">
+        Built with ❤️ by <a href="https://thyris.ai" target="_blank">Thyris.AI</a> | 
+        Licensed under Apache 2.0
+    </p>
+    """)
+    
+    # Event handlers
+    analyze_btn.click(
+        fn=analyze_text,
+        inputs=[input_text],
+        outputs=[redacted_output, detection_report, stats_output]
+    )
+    
+    input_text.submit(
+        fn=analyze_text,
+        inputs=[input_text],
+        outputs=[redacted_output, detection_report, stats_output]
+    )
+
+# Launch
+if __name__ == "__main__":
+    demo.launch()