Info
In short: these pages explain the web from the bottom up, in plain words, for people who automate browsers and need to know why a test behaves the way it does. Each page ends with the part that matters to seleniumbase-rs, so the explanation and the API sit side by side. Read them in order, or jump to the group that matches the problem you have.
Where to start#
| If a test… | Start with |
|---|---|
| passes sometimes and fails sometimes | How a page loads and renders, then JavaScript, events and the Web APIs |
| cannot find an element, or finds the wrong one | Selectors, the DOM and frameworks |
| is blocked, or meets a CAPTCHA | Detection and spoofing, then CAPTCHAs |
| loses its login, or needs to skip the login | Sessions, cookies, storage and logging in |
| fails only on CI or in a container | Containers, CI/CD and deployment, then Browser architecture |
| fails on one site after the site was deployed | Bundling, compression and caching |
| needs a proxy, a fake response or a recording of the traffic | The network, TCP, proxies and traffic capture |
| should be driven by an AI assistant | AI, machine learning and the Model Context Protocol |
Start here#
- How a page loads and renders: from a URL to bytes to pixels, what “loaded” really means, and why automation needs waits.
- How a modern browser is built: rendering and JavaScript engines, processes, the sandbox, targets, headless mode, profiles and launch flags.
Automating a browser#
- WebDriver and drivers: the W3C protocol, the driver binary, sessions, element references and why tests flake without waits.
- The DevTools Protocol and developer tools: the panels you use by hand and the WebSocket protocol behind them, which the pure CDP engine speaks directly.
- AI, machine learning and MCP: how an assistant drives a browser through the Model Context Protocol, and how to test sites that contain ML.
- Detection and spoofing: the signals a site reads to tell a program from a person, and where the library answers each one.
- CAPTCHAs: what they are for, how bot-management scoring works, and what the library does and does not do.
- Browser extensions: Manifest V3, content scripts, and how automation loads an extension.
Pages and the DOM#
- Selectors, the DOM and frameworks: the tree you point at, CSS and XPath, shadow DOM, iframes, what React, Vue and Angular leave behind, and stable selectors.
- JavaScript, events and the Web APIs: the event loop, trusted input, auto-waiting, observers, and calling browser APIs from a test.
Network and the wire#
- The network, TCP, proxies and traffic capture: sockets, HTTP, DNS, TLS, proxies, interception, HAR recording and what a connection reveals.
- Bundling, compression and caching: why the code a browser runs is not the code that was written, and why stale content breaks tests.
State, identity and forms#
- Sessions, cookies, storage and logging in: cookies, web storage, profiles, OAuth, JWTs, TOTP, passkeys, and saving a signed-in state.
- Forms: input types, validation, encodings, uploads, CSRF tokens and framework-controlled inputs.
Security#
- Web security and browser policies: the OWASP Top 10, the same-origin policy, CORS, CSP and the other rules a browser enforces, and what a test may check.
- Encryption and certificates: encoding, hashing and encryption told apart, signatures, the chain of trust and keeping secrets out of test code.
Building and shipping web apps#
- Frontends, backends, APIs and the shapes inside a server: what sits behind a page, from reverse proxies and REST to CRUD, middleware and serverless.
- Containers, CI/CD and deployment: Git, CI, Docker, Compose, deployment strategies and where browser tests run.
- Modern web technologies, eras and peer-to-peer: single-page apps, service workers, WebAssembly, WebRTC, HTTP/3, passkeys, Web3 wallets and P2P.
- Telemetry and observability: logs, metrics, traces, analytics scripts and the library’s own logging.
Reach: mobile, accessibility, SEO and media#
- The mobile web and accessibility: responsive design, device emulation, the accessibility tree and role selectors.
- Crawlers, spiders and SEO: crawl frontiers, robots.txt, politeness, and what search engines need from a page.
- Audio and video: codecs, adaptive streaming, autoplay, DRM and checking playback from tests.
Reference#
- Glossary: short definitions of every term used in these pages.