The fail-zero gate every project clears before it leaves our hands.
Site Checker is the internal QA plugin we built and maintain in-house. It runs across every WordPress engagement — rebuild, build, template customisation, refresh, redesign — before the partner agency or end-client ever sees the work. Nothing ships with an open Site Checker failure.
The first audience for the report is us. We run Site Checker on every site we touch — not because the partner agency requested it, but because we’d rather catch our own mistakes than have a client catch them. By the time the work reaches your QA layer, our pre-handoff gate has already closed. Whatever your agency surfaces post-handoff, we own the fix loop.
Five categories. Every page on the site, every environment.
Categories are simplified here for the public page; the internal taxonomy is finer-grained. Each surfaces a structured fail / warn / pass status against a known set of expected states.
-
CATEGORY 01
Core WordPress settings
WordPress core version is current; the search-engine indexing flag matches the environment (dev / staging / production); site title is configured; no leftover demo or test user accounts; debug logging and error display are not exposing the database to a public response.
-
CATEGORY 02
Content & SEO surface
Every page carries a unique meta title and description within length budget; one H1 per page with a clean H2/H3 hierarchy; no «Hello World», lorem ipsum, or default WordPress sample content; sitemap.xml exists, validates, and contains the URLs we expect.
-
CATEGORY 03
URL structure
Slugs are lowercase, hyphenated, free of post-IDs and query strings, free of stray non-Latin characters.
-
CATEGORY 04
Contact & integration
Phone numbers on the site match the approved list (no leftover dev placeholders); address links resolve to the correct map location; page titles follow the agreed capitalization style.
-
CATEGORY 05
Rebuild parityWhen the engagement is a rebuild
The original site is crawled and snapshotted before work starts. Site Checker compares the rebuild against that snapshot — missing copy, broken internal links, structural drift — and surfaces every divergence as a fail. Dormant on greenfield builds; live on rebuild engagements where an original site exists to compare against.
The output lands in the admin UI as a screenshot map — a side-by-side wall of every page as it actually renders, not as the dev team imagines it renders. One page, seen the way every device sees it, in the same review pass as the structured checks.
Responsive-only defects surface here, not in a follow-up bug ticket from the agency: overflowing text, broken hero ratios, navigation collapse, modal positioning, sticky headers that overlap content at narrower widths. We catch these at the gate, on every page, on every shipped resolution.
On rebuild engagements, the same capture pipeline runs against the original site before work starts. Post-build comparison against that snapshot is therefore visual and structured — we can show the agency, side by side, that the rebuild matches the original at every breakpoint they care about.
Across every WordPress engagement we run today, zero handoffs go out with an open Site Checker failure.
That isn’t a marketing line — it’s the operational rule. A red status on any check is a hard gate; the work doesn’t leave staging until the check passes or the deviation is documented and signed off. Warnings get triaged but don’t block. Passes are logged with timestamp and run-ID so the run history is auditable.
The gate exists because the alternative — trusting the dev team’s memory and a hand-written checklist — produced misses. A plugin that reads the database directly and re-renders the public site doesn’t get tired, doesn’t forget, doesn’t skip Friday afternoon.
Every case study under /work/ has an «Operational Integrity at handoff» section — the engagement-specific record of what cleared this gate before the work shipped.
If your agency wants to see the report itself — the structured pass / fail per category, with timestamps — we’ll export it on request. The report is share-friendly: no internal commentary, no source-code references, no naming of the partner agency. White-label by construction.
Each shipped version is tagged; every run records which version produced the result, so a report from six months ago can be reproduced or compared against the current run.
The plugin is internal — not on the WordPress.org plugin directory. The current version, the codebase, and the check-by-check ruleset are available for inspection on request. If your agency is evaluating us as a white-label partner and wants to walk the QA layer before signing scope, we welcome that conversation.
Have a project that needs a fail-zero handoff?
Send the Figma, the URL, or the spec. We’ll come back within 24 hours with a fixed-hours quote and a date. The Site Checker run is part of every engagement — not an add-on, not a line item, just how the work leaves our hands.