We built an agent that combines deep technical and contextual analysis of any text (emails, messages, URLs) and real-time fact-checking for optimal phishing, scam and ransomware protection. My co-founder Dylan and I (Kan) spent months reading about phishing, cybersecurity and scams, gathering a substantial amount of both real-life and synthetic data to power our original RAG-based system. Originally, the main challenge we encountered false positives so decided to redesign the entire system to be agentic, add real-time fact checking (crucial for credibility), reasoning and increase our knowledge-base 55x with more aggressive, vulgar but non-phishing examples, improve system prompt and bingo now we have a super accurate detection engine that almost never misses a phish or scam and never misclassifies a legitimate input as phishing. A really fun journey, now we're going hard on getting customers. We know phishing is the number one cause of data breach and ransomware for businesses and a lot of people get scammed every day so we're confident with dedication and consistency we'll get our first customer, and many more. We're particularly focusing on financial, legal and medical services, especially the SMEs in these sectors because we know these are the ones most targeted by phishing, data breach and ransomware while they mostly have no dedicated cybersecurity or threat prevention teams and are at the same time legally required to protect clients/patients data. Anyone wants to try or recommend VerifiedX we have a free tier so you can give it a spin - Just paste anything and our agent tells with the best accuracy if it's legit or not and why. app.verifiedx.me Kan