Cybersecurity β’ Artificial Intelligence β’ Full-Stack Development
I build secure, intelligent systems β offensive security agents, 3D network twin mapping, compliance analysis, and scam detection for Ghana.
Learn. Build. Secure. Repeat.
A cybersecurity awareness platform built around safe, defensive learning. Combines an authenticated Threat Academy, an AI Scam Analyzer, the CyberSENSE Assistant, interactive games, weekly competition flows, and superadmin reporting so learners build instincts through guided practice instead of passive reading.
Next.js App Router Β· TypeScript Β· Tailwind Β· Framer Motion Β· PostgreSQL Β· OpenRouter Β· OTP auth
A compliance document analyzer. Upload a policy or security document and receive a structured compliance assessment with severity-graded findings, regulatory references, and CVE enrichment.
A deterministic rule engine cross-references findings against a local knowledge base of cybersecurity frameworks, enriches technical vulnerabilities from the NVD, and generates an explanatory AI report β all without letting AI make any compliance decisions. Critical = 30, High = 20, Medium = 10, Low = 5; the score is clamped 0β100 and never calculated by a model.
Python 3.12 Β· FastAPI Β· Next.js 15 Β· TypeScript Β· Tailwind CSS v4 Β· OpenRouter Β· NVD API
**AI Network Twin.** Scan with Nmap, explore a live 3D graph of your network: hosts, operating systems, vulnerabilities, trust relationships, and animated attack paths β with an AI copilot that answers questions about the map.
Google Maps for corporate networks.
Nmap Β· FastAPI Β· React + Three.js Β· WebGL via React Three Fiber Β· Neon Postgres Β· WebSocket Β· OpenRouter
Ghana scam SMS detector β spot mobile money fraud before it hurts you. Paste any SMS, phone number, or MoMo reference to check if it's a scam.
Analyzes messages, Ghanaian numbers, MoMo transaction references, links, and screenshots. Ships with deep dives on Ghana's 8 deadliest tricks, real-world quizzes, and a community reporting dashboard. No accounts. No cookies. No tracking.
Accra Technical University Β· Project ATU 302
Next.js 16 Β· TypeScript Β· Tailwind CSS Β· OpenRouter (text + vision) Β· Neon Postgres Β· Vercel
Local AI-assisted media downloader. Paste a link or describe what you want; it extracts, ranks, normalises filenames, and files everything into a searchable library.
Two ways in β drop any supported URL and it detects the platform and downloads, or
describe the media in plain English and it parses intent, queries YouTube, ranks
candidates with an LLM, and normalises the filename to Artist - Title.mp4. The AI
does exactly three jobs: intent parsing, ranking, and normalisation.
Next.js 16 Β· React 19 Β· Tailwind CSS v4 Β· Neon Postgres Β· OpenRouter Β· yt-dlp Β· gallery-dl Β· ffmpeg
Work in progress, not yet public:
- HoneybyteAI β Windows-first personal AI blending chat with agentic automation
- ByteBite β campus food delivery platform, full-stack
- TECXE β local-first offensive security AI agent
- FRENSICS β digital forensics acquisition and recovery agent
|
AI-Powered Security Tools Agents that pair language models with reconnaissance, threat intelligence, and vulnerability analysis β with deterministic engines where the audit trail matters more than the prose. |
Digital Forensics & Investigations Local-first acquisition and recovery, deleted-file carving, chain-of-custody hashing, and AI-assisted reporting. |
|
Threat Intelligence & Automation Recon workflows, CVE enrichment from NVD, VirusTotal lookups, GeoIP, and automated PDF reporting. |
Linux & Web Application Security Nmap and network fingerprinting, OWASP-style web checks, and secure-by-default service design. |
|
3D & Real-Time Visualization WebGL network graphs, rotating disk visualisations, and streaming interfaces that make live state legible at a glance. |
|
Languages
Frameworks
Security & Forensics
AI
Media Tooling
Data & Infrastructure
Contributions are welcome. For anything security-related, only test systems you own or are explicitly authorized to assess.
Released under the MIT License.