How do I scan a page or repository for free?

The free scanner checks any public web page or GitHub repository for accessibility issues. No account is needed, and every report gets a permanent link you can share with your team.

Run a scan

Go to app.accesslint.com/scan.
Enter a page URL (like example.com/pricing) or a repository (like github.com/owner/repo).
Click Scan.

Page scans

For a URL, AccessLint loads the page in a real browser and runs its automated WCAG checks against the rendered result. Each issue in the report shows the rule, its impact (critical, serious, moderate, or minor), the WCAG level, the affected elements with an HTML snippet, and a link explaining the rule.

A few boundaries:

Only public pages can be scanned: nothing behind a login. To audit journeys that require sign-in, use Flows.
The scanner respects robots.txt: if a site asks automated tools not to fetch a page, AccessLint doesn’t.
Scans of the same site are rate-limited, so a burst of requests may be asked to wait.

Repository scans

For a GitHub repository, AccessLint fetches the default branch and runs static checks on its UI source files, the same file types covered in pull request reviews. Static checks catch what’s visible in the markup: missing alt text, unlabeled controls, missing landmarks, and similar.

Public repositories work with no setup. For a private repository, install the AccessLint GitHub App. Installing is what grants AccessLint permission to read it.

What a clean scan means (and doesn’t)

Automated checks find the issues a machine can detect: roughly a third to half of WCAG success criteria. They can’t judge keyboard flows, focus order, the quality of alternate text, or whether content makes sense to a screen reader user. Likewise, static repo checks can’t see issues that only appear on a rendered page, like insufficient color contrast. A clean scan is a good start, not a compliance verdict.

A clean scan is still a real result, though, and not a sign that something went wrong. When AccessLint reports no issues, the page loaded and every rule ran against it. If any part of a check has to be dropped for time, the scan says so rather than staying quiet about it.

One thing worth checking either way: a URL that redirects, or one that doesn’t exist, is scanned as whatever page the browser lands on. A clean report for a missing page is a clean report for that site’s “page not found” page. Confirm the page you scanned is the page you meant.

For continuous coverage instead of one-off scans, AccessLint reviews every pull request and monitors live journeys on a schedule.