Website Accessibility & SEO Auditor

Audit rendered webpages for WCAG accessibility defects and practical on-page SEO problems in one bounded crawl. The Actor returns page-level evidence that developers, agencies, and QA teams can export to JSON, CSV, Excel, or an API workflow.

What this website auditor checks

• WCAG 2.0 A and AA rules with axe-core against the rendered page
• affected CSS selectors, impact levels, failure summaries, and remediation links
• titles, meta descriptions, H1 usage, canonical links, image alt attributes, robots metadata, internal links, and JSON-LD presence
• accessibility and SEO scores, plus a prioritized cross-page issue summary
• optional full-page screenshots stored with stable collision-resistant keys

Best use cases

• pre-release accessibility QA for public landing pages
• agency website audits and technical prospect qualification
• recurring website quality checks with Apify Schedules
• structured issue exports for dashboards, spreadsheets, and engineering backlogs
• small JavaScript-rendered site crawls where raw HTML checks miss page content

Quick start

1. Add one or more public webpage URLs.
2. Start with one page and crawl depth zero.
3. Increase the page cap only after reviewing the first result.
4. Export the default dataset or connect it to your workflow.

Example input:

{
  "startUrls": [{ "url": "https://example.com" }],
  "maxPagesPerSite": 1,
  "maxDepth": 0,
  "respectRobotsTxt": true,
  "includeScreenshots": false
}

Input limits and cost control

startUrls accepts up to 10 public HTTP(S) URLs. maxPagesPerSite is capped at 50 and is the maximum number of billable successful page-audit results per starting site. maxDepth is capped at 5. The Actor respects robots.txt by default.

The live Pricing tab is authoritative. Under pay-per-event pricing, a successful saved page result is the page-audited event; invalid input, blocked private destinations, and failed page audits are not charged as page-audited events. User charge limits are respected and stop the crawl cleanly.

Output

Each successful dataset item contains the audited URL, title, timestamp, accessibility and SEO scores, issue count, issue details, link counts, and optional screenshot key. Failed page requests are preserved as uncharged diagnostic rows. The OUTPUT key-value record contains audited and failed page counts, average scores, total issues, and the most common issue IDs.

Security and responsible use

Only anonymously reachable public HTTP(S) pages are supported. The Actor rejects credential-bearing URLs, local hostnames, private/link-local/reserved IP destinations, and unsafe redirect or subresource requests. It does not accept cookies, passwords, or authenticated sessions and does not bypass CAPTCHAs or access controls.

Audit only websites you are authorized to test. This Actor is an automated QA aid, not a legal certification of WCAG, ADA, EAA, Section 508, or other compliance. Automated checks cannot replace expert manual review.

Reliability and known scope

One page failure does not erase successful results from the same run. The crawler is intentionally bounded and follows same-origin links only. Scores are product heuristics, not Google rankings or legal compliance scores. This release does not compare two runs automatically and does not generate PDF or SARIF reports.

Support

For bugs or feature requests, use the Actor's Issues tab and include the public test URL, run ID, and expected behavior. Do not include passwords, private URLs, or personal data.