app.py

import gradio as gr
import requests
import json
from bs4 import BeautifulSoup

def validate_schema(url):
    try:
        response = requests.get(url, timeout=5)
        response.raise_for_status()
        soup = BeautifulSoup(response.text, 'html.parser')
        scripts = soup.find_all('script', type='application/ld+json')
        results = []

        for script in scripts:
            try:
                data = json.loads(script.string)
                if isinstance(data, dict):
                    results.append(data)
                elif isinstance(data, list):
                    results.extend(data)
            except Exception as e:
                continue

        detected_types = list({item.get('@type', 'Unknown') for item in results if '@type' in item})
        core_types = {'Organization', 'Article', 'Person'}
        trust_trail = any(t in core_types for t in detected_types)
        schema_health = round(len(detected_types) / max(len(scripts), 1), 2)

        return {
            "trustTrail": trust_trail,
            "schemaHealth": schema_health,
            "detectedTypes": detected_types,
            "warnings": [] if trust_trail else ["No core Trust Trail types (Organization, Article, Person) found."]
        }

    except Exception as e:
        return {
            "trustTrail": False,
            "schemaHealth": 0.0,
            "detectedTypes": [],
            "warnings": [f"Error: {str(e)}"]
        }

demo = gr.Interface(
    fn=validate_schema,
    inputs=gr.Textbox(label="Enter URL to validate", placeholder="https://your-site.com/page"),
    outputs=gr.JSON(label="Validation Output"),
    title="AI Citation SEO – Trust Trail Validator",
    description="This tool checks for Trust Trail schema markup in a given URL and returns validation results."
)

if __name__ == "__main__":
    demo.launch()