Cybersecurity Glossary
Every term on HMMNM Security, explained in plain English. No prior knowledge needed — that is the point.
Malware
Any software built to harm you — viruses, ransomware, spyware are all malware. Think of it as the umbrella word for "bad program".
Phishing
Fake messages (emails, SMS, chats) that pretend to be someone you trust so you click a bad link or hand over a password.
Ransomware
Malware that encrypts your files and demands payment for the key. Backup offline, and you never need to pay.
Social engineering
Hacking humans instead of computers — tricking people into bypassing security by trusting, rushing, or fearing.
Zero-day
A flaw the vendor has had zero days to fix because nobody knew about it. Extremely valuable to attackers, scary for defenders.
Vulnerability
A weakness in software or configuration that an attacker can abuse. Every vulnerability is a bug — not every bug is a vulnerability.
Exploit
Code or a technique that actually takes advantage of a vulnerability. The vuln is the unlocked window; the exploit is climbing through it.
Payload
The part of an attack that does the real damage after getting in — stealing data, encrypting files, opening a backdoor.
Patch
A vendor fix for a vulnerability. "Patch Tuesday" exists because applying fixes fast is half of all security.
CVE
A public ID for a known vulnerability (CVE-2024-1234). One global catalogue so everyone can talk about the same flaw.
CVSS
A 0–10 severity score for CVEs. Rough guide: 9+ is critical, but a "low" score on your internet-facing server can still ruin your week.
Attack surface
Everything an attacker could touch: servers, apps, accounts, employees, even the physical office. Security shrinks it.
Threat model
A plain answer to: who might attack me, what do they want, and what am I protecting? You already do this when you lock your door.
Zero Trust
A model where no device or user is trusted by default — every request must prove itself, even from inside the network.
XSS (Cross-Site Scripting)
Injecting your JavaScript into a website so it runs in other visitors' browsers — stealing sessions or defacing pages. The classic web bug.
SQL injection
Typing database commands into input fields (like a login box) so the app runs them. Prevented with parameterized queries.
CSRF
Tricking your browser into sending a request you did not intend (like a money transfer) while you are still logged in. Fixed with anti-CSRF tokens.
RCE (Remote Code Execution)
The worst-case web bug: the attacker runs their own code on your server from the internet. Game over if unpatched.
SSRF
Making a server send requests to places it should not — often to internal services that "assume the inside is safe".
HTTP request smuggling
Confusing front-end and back-end servers about where one request ends, so a smuggled second request slips past security.
SSTI
Injecting code into template engines (the things that fill templates with data) so your input becomes server-side code.
IDOR
Changing a number in a URL (/invoice/1042 → /invoice/1043) and getting someone else's data because the server forgot to check ownership.
API
How programs talk to each other over the web. APIs are attacked the same as websites — plus juicy "hidden" endpoints.
Firewall
A filter between networks that allows or blocks traffic by rules. Necessary, but useless against tricks that look like normal traffic.
VPN
An encrypted tunnel that makes remote networks behave like they are local, and hides your traffic from the coffee-shop Wi-Fi.
Encryption
Scrambling data so only someone with the key can read it. Protects data in transit (TLS) and at rest (disk encryption).
Hashing
A one-way fingerprint of data. Passwords should be hashed, never encrypted — you verify, you never decrypt.
2FA / MFA
A second proof of identity (code, app, key) beyond the password. Stops most account takeovers even when the password leaks.
MITM (Man-in-the-Middle)
An attacker silently sitting between you and a server, reading or changing traffic. TLS certificates are the antidote.
DDoS
Flooding a service with traffic from thousands of machines until real users cannot get in. Availability attack, not a break-in.
Botnet
A herd of infected devices (bot = victim machine) remotely controlled as one — the muscle behind DDoS and spam.
C2 (Command & Control)
The attacker's channel for commanding infected machines. Cutting C2 can neutralize an intrusion without touching the malware.
Port scanning
Knocking on every door of a server to see which services answer. First step of almost every external pentest.
Penetration testing
Authorized, simulated attacking to find weaknesses before criminals do. Written rules and permission are what separate it from crime.
Red team / Blue team
Red = the attackers in the exercise. Blue = the defenders. Purple = both learning together.
Bug bounty
Companies paying researchers for responsibly reported vulnerabilities. A legal playground and a real career path.
EDR
Endpoint Detection & Response — software that watches every laptop/server for suspicious behaviour and can quarantine automatically.
Sandbox
A sealed, disposable environment for running suspicious files safely. If it is ransomware, better it encrypts a fake computer.
Honeypot
A decoy server that exists only to be attacked — every touch is a signal, and attackers waste time on it.
IoC (Indicator of Compromise)
Fingerprints of an attack: a bad IP, file hash, or domain you hunt for across your logs.
MITRE ATT&CK
A public catalogue of attacker tactics and techniques. The shared language defenders use to map and improve coverage.
Logs
The diary of everything systems did. Boring until the incident — then they are the only witness.
Backup rule (3-2-1)
Three copies, two different media, one offline/offsite. The only reliable ransomware insurance.
CTF
Capture The Flag — security puzzles in a competition format. The fastest legal way to practice offensive skills.
OSINT
Open-Source Intelligence — reconnaissance using only public info: job posts, LinkedIn, certificate logs, archived pages.
Responsible disclosure
Reporting a vulnerability privately so the vendor can fix it before it goes public. The ethical default for researchers.
Privilege escalation
Going from a small foothold (guest user) to full control (admin/root). Local (LPE) means on a machine you already reached.
