Inspiration
As a cybersecurity student pursuing CEH, I noticed that analyzing network logs, scan results, and security alerts requires strong experience and a lot of time. Beginners often struggle to understand what an alert actually means and what action should be taken. This inspired me to build CyberSleuth AI, an intelligent cyber investigation assistant powered by Gemini 3.
What it does
CyberSleuth AI helps users investigate cybersecurity incidents by analyzing logs, screenshots, scan outputs, and suspicious messages. The system identifies possible threats such as malware activity, phishing attempts, open ports, or misconfigurations and explains them in simple language along with recommended mitigation steps.
How I built it
The project was built as a web-based application integrated with the Gemini 3 API using Google AI Studio. Gemini 3’s advanced reasoning and multimodal understanding are used to process both text and image inputs. The AI reasons step-by-step like a cybersecurity analyst and generates structured investigation reports in real time.
Challenges I ran into
One of the biggest challenges was designing prompts that guide the AI to analyze security data accurately instead of giving generic responses. Balancing technical depth with beginner-friendly explanations was also challenging.
What I learned
Through this project, I learned how powerful multimodal AI can be in cybersecurity applications. I gained hands-on experience integrating Gemini 3, designing reasoning-focused prompts, and building an end-to-end AI-powered security tool.
What's next
Future improvements include real-time log ingestion, integration with SIEM tools, automated risk scoring, and support for enterprise-level incident response workflows.
Log in or sign up for Devpost to join the conversation.