Sentri Scan

• 

Inspiration

Every day, people download files, open links, and copy snippets of code without knowing whether they’re safe. Security tools are either too complex or too expensive, and most users never use them. We wanted to build a tool that makes file safety as simple as clicking “Scan” — fast, visual, and powered by AI-like intelligence, even without a real malware engine.

SentriScan was born from this idea: A futuristic, accessible threat-scanning experience for everyone.

What it does

SentriScan is an AI-simulated File Safety & Threat Intelligence Scanner. Users can:

Upload or simulate files, links, or script snippets

Get instant verdicts: Safe, Suspicious, or Dangerous

See risk scores, confidence levels, and human-readable explanations

Save reports, view scan history, adjust custom rules, and export JSON/Markdown/PDF

Use a neon cyberpunk security dashboard with animations and charts

Everything behaves like a real security SaaS — but powered by deterministic heuristics and dummy datasets.

How we built it

We built the project using:

React + TypeScript for the front-end

TailwindCSS & ShadCN for UI components

Framer Motion for futuristic cyberpunk animations

Recharts for visualizing threat data

A simulated backend API using deterministic rule-based scoring

Local JSON datasets for signatures, reputation, and policies

Fully responsive layout + neon/glassmorphism theme system

The scanning engine is a lightweight rule system that checks patterns like:

suspicious keywords (powershell, eval(, macro)

risky sources (USB, unknown links)

executable flags

URL reputation (simulated)

Scores map to verdicts that feel real and explainable.

Challenges we ran into

Designing a UI that feels like a futuristic security console without overwhelming the user

Ensuring all scans behave deterministically, even with simulated randomness

Building meaningful explanations for threats while using only dummy datasets

Making the app feel like a real SaaS product, not a prototype

Time constraints — building multiple full pages, animations, and data models quickly

Accomplishments that we're proud of

A fully navigable, production-like SaaS interface

Realistic scanning flow with progress animation + logs

High-quality neon cyberpunk UI with glassmorphism

Robust simulated engine with consistent, believable results

Threat Reports with filters, export features, and detailed drill-down

A complete end-to-end experience: Landing → Scan → Reports → Settings

The final product looks and feels like something teams could actually use.

What we learned

How to design a security tool that balances usability and depth

How to simulate complex AI/AV behavior with deterministic rules

How to build a full multi-page system quickly using component libraries

The importance of good UX during high-stress flows like scanning or errors

How visuals and motion can elevate the seriousness of a technical product

What's next for Sentri Scan

We want to extend SentriScan into a real security tool with:

VirusTotal / AbuseIPDB / URLScan integrations

LLM-powered static analysis for smarter explanations

Sandbox behavioral analysis

Team workspace mode (multi-user + alerts + logs)

Offline AI scanning engine for local file privacy

A full API for security teams to integrate SentriScan into workflows

The goal: Turn this demo into a lightweight, real-world AI safety scanner.

Built With

• api
• express.js
• framer-motion
• node.js
• react
• recharts
• routes
• serverless
• shadcn-ui
• supabase-/-local-json-storage
• tailwindcss
• typescript
• vercel