Cybersecurity access and attack terms are the vocabulary used by banks, payment firms and fraud teams to describe how systems are reached, watched and attacked. They fall into three groups. Access technologies such as remote desktop, secure shell, single sign-on and privileged access management decide who can log in and with what rights. Monitoring tools such as extended detection and response and behaviour analytics watch for misuse. Web attacks such as SQL injection and cross-site scripting exploit flaws in applications that handle money and card data.
The groups are linked in fraud cases. An attacker commonly gains entry through an exposed access service or a stolen credential, moves through the network using over-privileged accounts, and is either caught or missed by monitoring. Web flaws offer a separate route straight to stored data. Each member below is explained on its own.
Remote Desktop Protocol (RDP)
Remote Desktop Protocol is a Microsoft protocol that lets a user see and control a Windows computer from another machine over a network, by default through TCP port 3389. Support staff, administrators and remote workers use it daily, which is why it is so widespread.
An RDP service reachable from the internet is a frequent entry point for ransomware and account takeover, because attackers scan for open services and then try stolen or guessed passwords, a pattern seen in credential stuffing. Common safeguards are keeping RDP off the public internet, placing it behind a virtual private network, requiring multi-factor authentication, enabling network level authentication, limiting which accounts may connect and logging every session.
Secure Shell (SSH)
Secure Shell is an encrypted protocol for logging into and managing servers remotely and for transferring files, normally on port 22. It replaced older unencrypted tools and is the standard way administrators reach the Linux and network systems behind payment platforms.
Users authenticate either with a password or with a key pair: a private key held by the user and a matching public key stored on the server. Keys are stronger than passwords, but a stolen or poorly stored private key gives the thief the same access as its owner, often without a second factor. Weak points include keys left on shared machines, keys that never expire and forgotten keys of departed staff. Good practice is to protect keys with passphrases, rotate and remove them, restrict which hosts accept them and log all sessions.
Single Sign-On (SSO)
Single sign-on lets a user log in once to an identity provider and then reach many connected applications without separate passwords. The identity provider confirms who the user is and passes a signed assertion or token to each application, commonly through standards such as SAML or OpenID Connect.
For a bank or payment company, SSO gives central control: staff accounts can be created, reviewed and switched off in one place, and strong authentication can be enforced everywhere. The same centralisation concentrates risk, because a compromised SSO login, or a compromised identity provider, can open every connected system. Controls therefore include multi-factor authentication, short sessions, conditional rules based on device and location, and close monitoring of the identity provider.
Privileged Access Management (PAM)
Privileged access management is the set of tools and policies that control, record and limit accounts with administrator-level rights, such as database administrators and domain administrators. Typical features are a vault that stores and rotates privileged passwords, just-in-time access that grants rights only for a task and for a short time, and recording of sessions for later review.
Such accounts are the most valuable targets for attackers and for dishonest employees, so PAM supports control of insider fraud, where one person with broad access can move money or alter records. It also helps meet the access requirements of the PCI DSS, which expects access to cardholder data to be limited to those who need it and to be logged.
Extended Detection and Response (XDR)
Extended detection and response is a security platform that collects and correlates data from several sources, including endpoints, networks, email, cloud services and identity systems, to detect attacks and help staff respond. It grew out of endpoint detection and response, which watches individual devices, by widening the view across the environment.
It differs from a security information and event management system, which gathers logs from almost any source and relies heavily on rules written by the user, whereas XDR usually works most deeply with its own vendor's sensors and offers built-in response actions such as isolating a device. Payment firms use XDR to spot signs of intrusion, for example a staff laptop contacting an unknown server, before fraud losses occur. Many firms run both tools together.
User and Entity Behavior Analytics (UEBA)
User and entity behavior analytics applies statistics and machine learning to learn what is normal for each user and for each system, such as usual working hours, locations, data volumes and transaction patterns, and then flags departures from that baseline. Entities include servers, applications and devices as well as people.
It supports detection of insider threats and account takeover, where a legitimate login behaves unusually, for instance downloading customer files at night or approving payments in an unfamiliar sequence. Unlike a fixed rule, which catches only what its author anticipated, UEBA can surface unfamiliar patterns, although it produces false alarms until baselines settle. The approach is close to behavioral analytics applied to customers in fraud scoring.
SQL Injection (SQLi)
SQL injection is an attack in which an attacker enters specially crafted text into a form field, web address or other input so that the application passes it to its database as part of a command. If the application does not separate data from commands, the attacker can read, change or delete records, including card numbers and personal data. Large thefts of payment card data from retailers and processors in the late 2000s were reported to have involved this technique.
It is prevented mainly in code. Parameterised queries, also called prepared statements, treat input only as data, and input validation and least-privilege database accounts limit the damage. A web application firewall can block known attack patterns, and regular penetration testing checks for the flaw.
Cross-Site Scripting (XSS)
Cross-site scripting is an attack in which malicious script is made to run in the web browsers of other users of a trusted site. It takes three main forms: stored, where the script is saved on the site and served to every visitor; reflected, where it is carried in a link and bounced back by the server; and DOM-based, where the flaw lies in the page's own browser code.
On a checkout or login page the injected script can read what a customer types, steal session tokens or alter the page, a route to web-based card skimming. Defences include encoding output so that user text is shown rather than executed, validating input and setting a content security policy that restricts which scripts a page may run.
Ref 1140000220