Proxy & Scraping Glossary

An autonomous software program that uses AI to make decisions and complete tasks on its own, such as browsing the web or gathering data, with little human guidance.

A proxy that hides your real IP address from websites but still reveals that a proxy is being used.

A browser that lets you run many isolated profiles, each with its own fingerprint, cookies and proxy, so sites see them as separate genuine users.

Connecting two or more software systems through their APIs so they can share data and trigger actions automatically.

Pulling data from the JSON endpoints a website's own front-end already calls, instead of parsing its HTML — usually faster, cleaner, and more reliable.

A unique number that identifies a network on the internet — a block of IP addresses run by one organisation. It reveals whether an IP belongs to a home ISP, a data centre, or a mobile carrier.

Gradually increasing account activity on a new account to build history and appear legitimate before using it for target actions.

A single proxy endpoint that automatically rotates you through a large pool of IPs — you connect once, and the provider decides which address each request goes out from.

The amount of data that can be transferred over a connection in a given time — and the most common way proxy and VPN usage is measured and billed.

The set of techniques websites use to tell automated traffic apart from real human visitors, combining IP reputation, fingerprints, and behaviour signals.

Controlling a web browser through code instead of manual clicks, letting scripts navigate pages, fill forms, and extract data automatically.

Identifying you by combining dozens of details your browser reveals — screen size, fonts, GPU, timezone — into a signature unique enough to track you without cookies.

A security technique that runs web browsing in a separate, sandboxed environment so malicious code never touches the real device.

A saved set of browser data — cookies, history, and fingerprint settings — that isolates identities so each session looks like a distinct, consistent user.

Identifying your device by asking your browser to draw a hidden image. Tiny differences in your GPU, drivers, and fonts make the result unique — no cookies required.

A challenge–response test used to tell humans and bots apart, such as identifying images or checking a box, to block automated access.

A global network of servers that caches website content close to users for faster loading, while also filtering malicious requests and absorbing traffic spikes.

A technique carriers use to share one public IP among many customers — which is exactly why mobile proxy IPs are so trusted and hard to block.

A verification step Cloudflare shows to visitors it suspects are bots before allowing access to a site, ranging from an invisible JS check to an interactive puzzle.

Storing, sending, and rotating browser cookies so a scraper can maintain logins and sessions across requests, keeping automated traffic looking consistent and human.

The number of pages a search engine or crawler will fetch from a site within a given time, limiting how much of a large website gets crawled and indexed.

How fast a crawler sends requests to a website, usually measured in requests per second. Keeping it reasonable avoids overloading servers and triggering blocks.

The process of pulling specific information from a source, such as a web page, and turning it into structured, usable data — the core goal of web scraping.

An IP address hosted on servers in a data center rather than assigned by an ISP — offering high speed and low cost, but easier for websites to detect.

An IP address assigned exclusively to one user, so no one else shares the bandwidth or the IP's reputation.

The internet's phonebook — it translates human-readable domain names like example.com into the numeric IP addresses computers use to connect.

When your device sends domain lookups outside the VPN tunnel, so your ISP can still see every site you visit even though the traffic itself is encrypted.

A proxy that hides your IP address and sends no headers that reveal a proxy is being used, so the request looks like an ordinary direct connection.

A proxy that sits between you and the internet, makes requests on your behalf, and is the everyday kind of proxy used to change your IP or reach blocked content.

Selecting proxy IPs from a specific country, region, or city so your requests appear to originate from that exact location.

Restricting access to content based on the geographic location of the visitor's IP address.

A real browser running without a visible window. It loads pages and runs JavaScript exactly like Chrome does, but is driven by code rather than a mouse and keyboard.

A hidden element or decoy placed on a website to catch bots. Real users never interact with it, so any request that touches it is flagged as automated.

Small pieces of metadata sent with every web request and response, describing things like the browser, content type, and cookies — quietly shaping how servers treat your traffic.

An intermediary server that forwards web (HTTP/HTTPS) requests on your behalf, able to read, cache, and filter traffic at the application layer.

The second major version of HTTP that loads web pages faster by sending many requests over one connection at the same time, replacing HTTP/1.1's one-request-at-a-time model.

The newest version of HTTP that runs over the QUIC protocol on UDP instead of TCP, making connections faster and more resilient, especially on mobile and unstable networks.

The secure version of HTTP that encrypts data between your browser and a website using SSL/TLS, preventing eavesdroppers from reading or tampering with the traffic.

A fast, stable VPN protocol especially good at keeping your connection alive when you switch networks, commonly used on mobile devices.

A trust score that websites and security systems assign to an IP address based on its past behaviour — a clean IP is trusted, while a flagged one may be blocked or challenged.

The practice of automatically cycling through multiple IP addresses so successive requests appear to originate from different users.

A security method that only lets pre-approved IP addresses connect to a service, blocking everyone else by default.

The fourth and most widely used version of the Internet Protocol, giving every connected device a numeric address like 192.168.1.1, with about 4.3 billion unique addresses.

The newest version of the Internet Protocol, created to replace IPv4 because the world ran out of addresses, offering a virtually unlimited supply of unique IPs.

A datacenter-hosted IP that is registered to an Internet Service Provider, combining datacenter speed with residential-level trust.

When a website blocks all traffic from a specific IP address, refusing requests or returning error pages.

A hash that identifies a client based on how it sets up an encrypted TLS connection. Servers use it to recognise and block automated tools even when IP addresses change.

A VPN feature that cuts your internet connection the instant your VPN drops, so your real IP address is never exposed during the gap before reconnection.

An open standard that lets AI applications connect to external tools, data sources, and services through one consistent client-server interface.

A proxy that routes traffic through a real 4G or 5G cellular connection. Because carriers share one IP among thousands of phone users, these are the hardest proxies for websites to block.

A VPN that routes your traffic through two or more servers instead of one, so no single server ever knows both who you are and what you are visiting.

Operating multiple accounts on the same platform, requiring unique fingerprints and dedicated IPs per account to avoid bans.

A VPN or proxy provider's commitment not to record what you do online — your browsing, connections, traffic, or real IP.

A VPN that disguises its traffic to look like ordinary HTTPS web traffic, helping you bypass firewalls and censorship that block standard VPN connections.

A popular open-source VPN protocol that creates a secure, encrypted tunnel between your device and a server, trusted for its high security and wide platform support.

An open-source browser automation framework from Microsoft that controls Chromium, Firefox, and WebKit — widely used for testing and JavaScript-heavy web scraping.

A numeric label that directs network traffic to the right program on a device. Combined with an IP address, it tells data exactly where to go.

An IP address dedicated to a single user, giving you exclusive use, full bandwidth, and a reputation no one else can spoil.

The process of proving you are allowed to use a proxy, usually with a username and password or by pre-approving your IP address.

A collection of many proxy IP addresses that a tool rotates through to spread requests and avoid getting blocked.

A middleman between you and the internet. Your request goes to the proxy first, and the proxy forwards it — so the website sees the proxy's IP address instead of yours.

A Node.js library by Google that controls Chrome and Chromium through the DevTools Protocol, popular for scraping JavaScript-rendered pages and automating Chrome tasks.

Routing traffic through multiple proxies in sequence so each one only knows the previous hop, not the original source or final destination.

A modern transport protocol that sends data over UDP with built-in encryption and faster connection setup — the foundation HTTP/3 is built on.

A restriction on how many requests a client can make in a given time — one of the most common defences scrapers must work around.

A browser that runs in the cloud or on a remote server and is controlled over the network, used for scalable automation, scraping, and isolation.

Deliberately slowing down how often a scraper sends requests to stay under a site's limits, helping avoid bans, server overload, and detection.

A proxy that routes your traffic through a real home internet connection, so websites see an ordinary ISP customer rather than a server in a data centre.

A proxy that sits in front of web servers and answers requests on their behalf, handling load balancing, caching, HTTPS, and security.

A simple text file at the root of a website that tells crawlers which pages they may or may not access — a voluntary set of rules, not a technical lock.

A proxy that automatically assigns a different IP address from a pool for each request or on a set interval, spreading traffic to avoid blocks.

Software "bots" that mimic the repetitive clicks, typing, and steps a human performs in apps and websites, automating routine digital tasks.

A long-established browser automation framework that controls real browsers through the WebDriver standard, supporting many languages and widely used for testing and scraping.

A temporary cookie a website uses to remember you during a single visit, such as keeping you logged in, deleted when you close the browser or the session ends.

An open-source encrypted proxy designed to bypass internet censorship by disguising traffic to slip past firewalls that block traditional VPNs.

A proxy server whose IP address is used by multiple people at the same time, splitting the cost and bandwidth among everyone.

The latest SOCKS proxy protocol that routes any kind of network traffic — TCP and UDP — through a proxy with optional authentication and no application-layer awareness.

A VPN feature that routes some traffic through the encrypted VPN tunnel while other traffic goes directly to the internet, giving control over which apps use the VPN.

An SSH protocol tunnel that encrypts and forwards traffic between your device and a remote server — a simple way to secure connections or bypass network restrictions.

The encryption protocol that secures data in transit on the web — the S in HTTPS — protecting it from eavesdropping and tampering.

A proxy mode that keeps the same IP for a set period so multi-step workflows like logging in and checking out stay on one consistent address.

A smaller network carved out of a larger IP address range, grouping related addresses together for efficient organisation and routing.

Maintaining separate, isolated browser sessions for each account to prevent cross-contamination of cookies, fingerprints, and IP history.

The core set of rules that lets computers find each other and exchange data reliably across the internet, combining IP for addressing with TCP for ordered, error-checked delivery.

Identifying a client by analysing how it negotiates an encrypted connection, letting servers detect specific tools even before any web request data is exchanged.

A proxy that intercepts your traffic without hiding your IP address, and often without you even knowing it is there.

A fast, connectionless way to send data over the internet without guaranteeing delivery or order, trading reliability for low latency.

The identifying string a browser sends with every request, telling the server which browser, version, and operating system is making the request.

Encrypts all of your device's internet traffic and routes it through a remote server, hiding your IP and protecting data on untrusted networks.

The automated extraction of data from websites — fetching pages programmatically and parsing their content into structured data.

An automated message sent from one app to another the moment an event happens, instead of constantly polling for updates.

When your real IP address is exposed through the browser's built-in real-time communication feature, even when using a VPN or proxy.

A protocol that opens a persistent, two-way connection between a browser and a server, letting data flow in both directions continuously without re-connecting each time.

A modern VPN protocol built to be fast and simple — its code is small enough to audit properly, and it connects quicker and runs faster than older protocols.