Campbell Davis

Case Study — Vibe-Code Health Check

An AI-Built WordPress/WooCommerce Integration

An overhead view of hands ringing up a sale on a point-of-sale tablet in a small shop

35,000

lines of code

190

classes

4

priority actions

The situation

A business had an AI-generated WordPress plugin syncing product data from an ERP system into WooCommerce, and wanted to know if it was safe to rely on before treating it as production infrastructure.

The approach

  • Architecture review
  • Automated complexity/quality scanning
  • Security/standards check
  • Review of existing automated test coverage

What was found

A surprisingly large codebase (35,000 lines across 190 classes) for the problem it solved. That’s typical of AI-generated code, which favours duplication over abstraction and rarely cleans up dead code. Test coverage told a similar story: strong unit test coverage, but no functional or business-facing tests. Several security concerns were also flagged.

The recommendation

  • Keep the current architecture unless significant further feature growth is planned, in which case a more decoupled architecture was recommended.
  • Add human-readable functional tests so business requirements are captured in plain language, alongside the existing unit tests.
  • Adopt automated quality and standards scanning as an ongoing gate before each release.
  • Fix flagged security issues before any production release.

The outcome

The client walked away with clarity and a prioritised, cost-aware action plan — not a rewrite they didn’t need. That’s the approach of the Health Check and Vibe-to-Production offerings: honest assessment first, then targeted improvements to maximise impact.

Not sure where your own project stands?

A health check gives you a clear, honest picture before you rely on it.

Book a health check