How Pourscan works
From first launch to the final report: what the app does and what it does not.
Crawling
Pourscan starts from the site's sitemap and robots.txt, then can follow links from the home page. Every page is downloaded and screenshotted with Chromium in two viewports, desktop and mobile.
Checks
On the code: automated checks on structure, text alternatives, forms, tables, language, headings and ARIA. On the rendering: colour contrast, target size (WCAG 2.2), reflow at 320 pixels, off-screen content, visible focus.
Outcomes
Every criterion gets a technical outcome: satisfied, not satisfied, not applicable (the content is absent) or not verifiable automatically (a human review is needed, and the app says which). Notes that do not affect conformance are kept separate.
Reports and fixes
HTML and PDF reports with highlighted screenshots, a fix package for developers, scan comparison and per-URL history.
What it does not do
It does not replace human evaluation or the accessibility statement: it prepares them. It never sends the analysed site's content to the server.