if exposures: return "feature_value": len(exposures), "risk_level": "CRITICAL" if any("private_key" in e["exposed_items"] for e in exposures) else "HIGH", "details": exposures else: return "feature_value": 1, # mention exists but no high-risk secrets "risk_level": "MEDIUM", "details": ["paste_url": p["url"], "date": p["date"] for p in pastes] feature_output = generate_feature() print(feature_output) Example Feature Output "feature_value": 2, "risk_level": "HIGH", "details": [ "paste_url": "https://pastebin.com/abc123", "date": "2026-04-04", "exposed_items": "api_key": ["sk_live_4eC39HqLyjWDarjtT1zdp7dc"] , "paste_url": "https://pastebin.com/xyz789", "date": "2026-02-22", "exposed_items": "password": ["KeyInv@2024"] ]
# Simulated response: list of pastes that match mock_pastes = [ "title": "investment_api_keys.txt", "content_snippet": "KEY_INVESTMENT_API_KEY = 'sk_live_4eC39HqLyjWDarjtT1zdp7dc'", "date": datetime.now() - timedelta(days=10), "url": "https://pastebin.com/abc123" , "title": "config_backup", "content_snippet": "DB_PASSWORD=KeyInv@2024", "date": datetime.now() - timedelta(days=45), "url": "https://pastebin.com/xyz789" ] site%3apastebin.com+key+investment+services
findings = {} for label, regex in patterns.items(): matches = re.findall(regex, text, re.IGNORECASE) if matches: findings[label] = matches return findings def generate_feature(company_name="Key Investment Services"): pastes = fetch_pastebin_mentions(company_name) if exposures: return "feature_value": len(exposures)
if not pastes: return "feature_value": 0, "risk_level": "LOW", "details": "No mentions found on Pastebin." "details": exposures else: return "feature_value": 1
return mock_pastes def extract_sensitive_patterns(text): patterns = "api_key": r'[A-Za-z0-9_]32,', "private_key": r'-----BEGIN (RSA
exposures = [] for paste in pastes: sensitive = extract_sensitive_patterns(paste["content_snippet"]) if sensitive: exposures.append( "paste_url": paste["url"], "date": paste["date"], "exposed_items": sensitive )