agent.py

import time
import re
import os
import tempfile 
from playwright.sync_api import sync_playwright

def clean_url(url: str) -> str:
    url = url.strip()
    url = re.sub(r'^ht{2,}ps', 'https', url)
    if not url.startswith(('http://', 'https://')):
        print(f"⚠️ Scout: Assuming HTTPS for {url}")
        return f"https://{url}"
    return url

def scrape_website(url: str):
    target_url = clean_url(url)
    print(f"🕵️  Scout Agent: Heading to {target_url}...")
    
    with sync_playwright() as p:
        browser = p.chromium.launch(
            headless=True, 
            args=["--disable-blink-features=AutomationControlled", "--no-sandbox"]
        )
        
        context = browser.new_context(
            viewport={"width": 1400, "height": 1000},
            user_agent="Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
        )
        
        page = context.new_page()
        
        try:
            page.goto(target_url, wait_until="domcontentloaded", timeout=60000)
            
            # Scroll Logic
            for i in range(5):
                page.mouse.wheel(0, 500)
                time.sleep(0.5)
            page.evaluate("window.scrollTo(0, document.body.scrollHeight)")
            time.sleep(1)
            page.evaluate("window.scrollTo(0, 0)")
            time.sleep(1)
            
            # --- PERMISSION FIX ---
            # We construct a path inside the system's temporary directory
            temp_dir = tempfile.gettempdir()
            screenshot_path = os.path.join(temp_dir, "screenshot_input.jpg")
            
            page.screenshot(
                path=screenshot_path, 
                full_page=True, 
                quality=80, 
                type='jpeg'
            )
            
            print(f"📸 Scout Agent: Evidence secured at {screenshot_path}")
            text_content = page.inner_text("body")
            
            return {
                "screenshot_path": screenshot_path,
                "text_content": text_content,
                "status": "success"
            }

        except Exception as e:
            error_msg = str(e)
            print(f"❌ Scout Agent Error: {error_msg}")
            if "ERR_ABORTED" in error_msg or "ERR_NAME_NOT_RESOLVED" in error_msg:
                return {"status": "error", "error": f"Could not reach {target_url}.", "screenshot_path": None}
            return {"status": "error", "error": f"Browser Error: {error_msg[:100]}...", "screenshot_path": None}
            
        finally:
            browser.close()

# Test block
if __name__ == "__main__":
    scrape_website("https://example.com")