Current version
The plugin is ready for external testing.
Version 0.1.0-rc.2 is reproducibly built and passes the internal gates. Public download is not open yet: the exact RC is first tested on owned and authorized sites, then goes through a separate external review and WordPress.org submission.
- Not a lead form. A complete basic report needs no email, account, or payment.
- Not an auto-fixer. The plugin analyses and explains but does not modify the audited site.
- Not a promise of a perfect site. Unavailable scope is marked not tested.
What it looks like
A real report, not a product mock-up.
This is an unretouched WordPress capture from the exact 0.1.0-rc.2 interface. The selected report language is English even when the administrator uses another dashboard locale.

Coverage, priorities, and next actions lead the report. Technical evidence expands only when it is useful.
Why it helps
One baseline before several specialist audits.
One local report helps identify whether the next person should be a developer, SEO practitioner, UX researcher, security specialist, or legal adviser before commissioning several overlapping initial audits.
01
WordPress and server
Versions, configuration, components, cron, database, caches, and accessible server signals.
02
SEO and frontend
Indexability, metadata, links, templates, rendered reflow, forms, and keyboard signals.
03
Security and privacy
Observable posture and data-flow signals with explicit manual and legal boundaries.
Installation
Install it like a regular WordPress plugin.
- After public release: open Plugins → Add New, find 320px Site Audit, and choose Install.
- For a controlled RC: if the 320px team supplied a ZIP, choose Upload Plugin, select the archive, and activate it.
- Then: open Tools → 320px Site Audit. No registration, licence key, or external dashboard connection is required.
Five profiles
Start with Quick. Go deeper only when needed.
The numbers are exact planned checks in this RC, not a promise that the plugin will find that many problems.
20
Quick
A broad local baseline with no network requests.
46
Full
An extended profile with a separate exact-origin crawl confirmation.
11
Deep code
Bounded custom/MU-code analysis without storing source.
11
Vendor
Inspect only 1–10 explicitly selected themes or plugins.
20
Rendered
A local browser view of DOM, layout, and accessibility after separate confirmation.
Reading the report
Three questions instead of one magic score.
- Coverage: what could be observed? Not tested is a method or access boundary, not a good result.
- Finding: what supports the conclusion? Each item has evidence, confidence, affected scope, and a safe next step.
- Priority: what comes first? Actions are ordered as now, next, and later without false percentage precision.
- Comparison: what changed? Only compatible completed runs with the same profile and method are compared.
Data boundary
By default, the report goes nowhere.
- Local scans, viewing, comparison, JSON/CSV, and print do not require a 320px account or connection.
- Scheduled jobs cannot send site URLs, inventory, reports, or personal data.
- Optional external evidence runs only after a screen names the recipient, purpose, exact data class, terms, and privacy policy.
- A review request defaults to contact only. Any report scope is prepared and masked locally, then shown in full before separate consent.
- Screenshots and the private browser bundle are never part of a review request.
Market reference
What separate initial audits cost in the reviewed sample.
Open Russian-market snapshot on 22 Aug 2026 for a small service site: technical audit median RUB 20,000 (N=6, range 10–30k); SEO audit RUB 30,000 (N=6, 25–30k); rendered/UX baseline RUB 20,000 (N=7, 10–150k).
Adding medians gives a roughly RUB 70,000 separate-order scenario, not guaranteed savings: scopes overlap and an automated baseline is narrower than manual expertise. Combined public offers in the same review started around RUB 45,000–50,000. Security/pentest and legal review are excluded.
Method
Why this is not an invented marketing number.
Only open RUB prices for an applicable format are included; “from” is treated as a lower bound, the median is central, and the full range is shown. This is not official market statistics. The 19 category offers were reviewed on 22 Aug 2026 and are rechecked before production and quarterly.
Does not replace
Decisions that need a person and context.
- SEO strategy, semantics, competitor work, or a ranking guarantee.
- UX research and user testing.
- Penetration tests, incident forensics, and manual code review.
- Legal audit, selection of legal grounds, documents, and regulator notification.
- Business priorities, remediation, and implementation ownership.
Tell us what the site needs.
A goal, the current difficulty, and an optional website address are enough. Do not send passwords, keys, or private exports.