Cybersecurity
Phishing URL Detection
Classify malicious links from their structure, not a blocklist
- Python
- scikit-learn
- Flask
- Pandas
Source code · Report · Viva prep
Security projects impress because they sit where networking, cryptography and machine learning meet. They are also where hand-waving gets caught quickly, since an examiner can ask exactly which attack you detect, how you measured false positives, or why you chose AES over RSA for a given step.
Ready to build
Cybersecurity
Classify malicious links from their structure, not a blocklist
Cybersecurity
Client-side AES encryption with hidden-payload image transport
Cybersecurity
Machine-learning anomaly detection over live network traffic
Cybersecurity
Flags spliced and copy-move edits using error level analysis and a CNN
Cybersecurity
Classifies abusive messages and queues them for human moderation
Cybersecurity
Behavioural biometrics that keep checking who is at the keyboard
Cybersecurity
Vault encrypted in the browser, synced as ciphertext, with autofill
Cybersecurity
Classifies Windows executables as benign or malicious from static features
Cybersecurity
Classifies query payloads and blocks malicious ones in web middleware
Cybersecurity
Crawls a lab web app and reports common web weaknesses
Cybersecurity
Fake SSH and HTTP services that log and map attacker behaviour
Cybersecurity
TOTP enrolment via QR, backup codes and rate-limited verification
Cybersecurity
Encrypt files in the browser, share expiring links, log every access
Cybersecurity
Ryu controller that classifies flows and installs mitigation rules
Cybersecurity
Flags risky APKs from their requested permissions and API calls
Cybersecurity
Ingests syslog and auth logs, correlates events and alerts on anomalies
Cybersecurity
Click-point and image-sequence login that resists shoulder surfing
Cybersecurity
Frame CNN plus a temporal model that flags manipulated faces
Cybersecurity
Classifies emails as phishing from their body text and headers
What you get
These kits come with a threat model, the datasets or captures used for testing and a report that documents the attacks covered and the ones that are out of scope. That last part is what makes a security project defensible.
Your topic is not in the catalogue yet? We build to your problem statement too.
Ask about a custom buildQuestions
Yes. Every kit is defensive, meaning it detects, encrypts or protects, and all testing runs against public datasets or your own machine. None of them attacks systems you do not own.
No. They run on Windows, macOS or Linux with Python installed. Some optional testing steps are easier on Linux, and the guide covers both.
Encryption and steganography projects are the most self-contained. Intrusion detection is more impressive, but it expects you to explain both networking basics and model evaluation.
Deadline approaching?
Send your abstract, rubric or guidelines document. You get a straight answer on scope, timeline and price within a day — no obligation.