WebAuditOS website audit platformOne URL. 9 categories.
Every finding shows its evidence.
199 checks across the whole site, one weighted score, and the header, URL or DOM node behind every finding. Not a verdict you have to take on trust.
Test your website now
Free, no card. We read only public pages and respect robots.txt.
Try Pro free for 7 days


199 checks across 9 categories
The same check registry on every plan. Each category links to its documentation.
Every site you look after. And what the last scan found on it.
Each project, its score, the pages behind it and the finding that needs you next.
Anything can return a number. We show the proof.
The response a finding came from, how sure we are of it, and what the run never measured.
Evidence under every finding
Each finding carries the observation behind it - the missing header, the URL that answered, the node that failed - and says whether we measured it, inferred it, or need you to confirm it. Each is mapped to its standard, so what you take to a client is not our opinion.
See it as a visitor will
Every crawled page captured at mobile, tablet and desktop widths, side by side and to scale. The mobile and desktop passes are the ones that score.
Scan it once, then keep scanning
Set a schedule and the audit repeats itself. Put two runs side by side to see what appeared, went away or got worse. Alerts go out only for a regression, and runs that are not comparable are refused rather than compared.
Every page the crawl touched
Each URL with its type, status, canonical, indexability and its own score, so the site-wide number always traces back to the pages that made it. The crawl is deterministic: scan twice, get the same pages in the same order.
Unmeasured, never “clean”
When the browser does not come up, the categories that depend on it are reported as unmeasured and the overall is recomputed over the rest. A narrower scan can never score better than a full one.
Numbers in each standard's words
Findings cite OWASP ASVS, WCAG 2.2, the RFC or the Core Web Vitals threshold they are measured against, with the affected URLs listed underneath.
Decisions that stick
Mark a finding fixed, not applicable or ignored and the decision is kept for the workspace - which is why check IDs are never renamed.
Built for the work that follows. Track it, hand it over, keep clients apart.
Page inventory
Every crawled URL, its score and its issues.
OpenPreview Lab
Page captures in device and browser frames.
OpenClient reports
Choose the sections, add your branding, export.
OpenMonitoring
Scheduled scans, and alerts only when something got worse.
OpenReadiness checklist
The server-side facts a public scan cannot verify, signed off by name.
OpenClient projects
A project per client site, and on any paid plan a share link for a client who should see only theirs.
OpenA scan is an observation, not an intrusion.
Everything below is enforced in the crawler, not promised in a policy - and stated in the same words on /security and in the terms.
Only what is already public
GET requests and nothing else: no forms, no credentials, no POST, nothing that changes anything on your side. Pages behind a login are not covered, because we send no credentials to anyone.
robots.txt decides, and 4xx means no
It is read before a URL is fetched rather than after, and a 401, 403, 429 or 5xx is treated as a refusal. That is deliberately stricter than Googlebot, which reads the same answers as permission.
One line takes you out
A rule naming WebAuditOSBot stops us on the next request. There is a public opt-out page and a blocklist as well, and stored evidence is deleted on the retention window of the plan that produced it, save that the newest scan of each of your own projects is kept beyond the window until you delete the project or the workspace, so a dashboard does not go blank.
The same API the app uses
Start scans, read findings and list pages from your own build, with a key per workspace and cursor pagination.
The week you never have to be present for.
A paid project keeps auditing itself. Each step below is what the engine actually does, in the order it does it, with numbers from the same run as the sample report.
All 9 categories on every plan. Only the limits change.
Free gives you the whole audit - every finding, the page inventory, screenshots and history. Paying buys crawl budget, retention, seats and the things you do with an audit afterwards.
Free
€0 forever
A real audit of a small site, free, with nothing held back.
- Scans a month
- 3 / month
- Pages a crawl
- 8 per crawl
- History kept
- 30 days
- Seats
- 1
- The full findings list, page inventory and all 9 category scores
- Screenshots and Core Web Vitals for every rendered page
Pro7 days free
€49 /mo
Deep audits with monitoring and history.
- Scans a month
- 300 / month
- Pages a crawl
- 1 000 per crawl
- History kept
- 12 months
- Seats
- 5
- Everything in Starter
- Daily monitoring with alerts
VAT included - the price is final
Start 7-day trialStarter
€19 /mo
Keep one product measurably healthy.
- Scans a month
- 15 / month
- Pages a crawl
- 100 per crawl
- History kept
- 60 days
- Seats
- 1
- Everything in Free, with a far larger crawl budget
- Weekly and monthly scheduled scans, with regression alerts by e-mail
VAT included - the price is final
Choose StarterSuper
€129 /mo
Measure yourself against the competition, not against last month.
- Scans a month
- 900 / month
- Pages a crawl
- 2 000 per crawl
- History kept
- 24 months
- Seats
- 8
- Everything in Pro
- Competitor benchmark - your site beside 5 rivals, scored on the same scale
VAT included - the price is final
Choose SuperAgency
€399 /mo
Client-ready reporting under your own brand.
- Scans a month
- 1 500 / month
- Pages a crawl
- 5 000 per crawl
- History kept
- 24 months
- Seats
- 15
- Everything in Super
- White-label reports - your logo and accent colour, our branding removed
VAT included - the price is final
Choose AgencyEnterprise
Custom
Governance, scale and procurement-ready terms.
- Scans a month
- Custom
- Pages a crawl
- Custom
- History kept
- Custom
- Seats
- Custom
- Everything in Agency, with the numeric limits removed
- Scans, seats, projects and history retention agreed per contract
Price and terms agreed per contract
Talk to salesEvery limit side by side, and the rows these cards leave out - scan comparison, readiness sign-off, API rate, support - are on the full table. Compare all 6 plans.
Questions, answered plainly. What people ask before they point this at a site.
Running a scan
Is it safe to run this against a live site?
Yes - a scan is a read-only observation of pages that are already public. We send GET requests only: no forms, no credentials, no POST, and nothing that changes anything on your side. robots.txt is checked before a URL is fetched rather than after. With no Crawl-delay declared we keep up to 4 requests in flight with at least 250 ms between requests in each of them, which is at most 16 requests a second; declare one and we honour it in full, at a single request in flight.
Read exactly what a scan does and does not doCan I audit a site I don't own?
Only one you own or are authorised to assess - you tick a box confirming exactly that when you start a scan, the confirmation is recorded, and it is in the terms. Auditing client domains is the main case this is built for, which is what the agency workspaces and white-label reports are for. If you are on the receiving end of a scan you did not ask for, one line in your robots.txt stops us on the next request.
Authorisation to scan, in the termsDoes it work on a JavaScript app or a site behind a login?
A JavaScript app, yes: pages are rendered in Chromium, so content that only appears after the JavaScript runs is seen and measured. On a large site the browser pass renders a sample of the crawled pages rather than all of them, and the scan tells you how many of each it did. Pages behind a login are not covered - we send no credentials to anyone, by design.
What the report gives you
How is this different from running Lighthouse or a free header checker?
Those answer one question, about one page, and leave you nothing to hand anybody. This crawls the site, runs 199 checks across 9 categories, and returns one weighted score with the evidence attached to every finding - plus a prioritised fix list and a report you can send to a client, branded on Agency plans. The difference is not the checks, it is having one defensible artefact instead of four browser tabs and a spreadsheet.
Compare it to your current processWhat can a scan not tell me?
Quite a lot, and we label it instead of guessing. Password hashing, CSRF tokens, rate limiting, row-level security and whether your backups restore are invisible from outside - no external scanner can see them, including this one. They sit in a manual attestation layer that never silently moves your score, and every finding is marked as auto-detected, inferred from public signals, or requiring your confirmation.
How the score is built, with the arithmeticPlans and access
Do I need an account or a card to try it?
Neither. Paste a URL on this page and you get a score across all 9 categories, your highest-priority finding in full, and a link you can share - with no sign-up. An account is what keeps the history and opens the rest of the findings. Pro is the one paid plan with a trial - 7 days, card required, cancel any time. The others bill from the first day and can also be cancelled at any time.
See what each plan addsUnderstand common website findings
Read the evidence, practical fixes and limits behind each check. Browse all website checks.
Who provides this service.
WebAuditOS is built and operated by an independent software company registered in the Czech Republic. We are the sole provider of this service and the controller of the data described in our privacy policy.
- Operating company
- ADMI ENTERPRISES s.r.o.
- Registered seat
- Nové sady 988/2, Staré Brno, 602 00 Brno
- Company ID (IČO)
- 29581010
- Commercial register
- C 151699, Krajský soud v Brně (Regional Court in Brno)
- VAT
- Sales are made through our merchant of record, Stripe (Link), which includes and remits any applicable VAT - listed prices are final, with nothing added at checkout.
- [email protected]
Audit the site before your client does.
One URL in. A report you can defend line by line out.



