Registering WordPress AI abilities with categories
Notes on registering custom WordPress AI abilities now that categories are required, with a CRM example and why categorizing keeps integrations manageable.
Notes on registering custom WordPress AI abilities now that categories are required, with a CRM example and why categorizing keeps integrations manageable.
Why heavily updated WordPress sites bloat their object cache, and how 6.9’s consistent cache keys for query groups fix it, with a WooCommerce case and sample code.
My honest read on the WordPress AI Abilities API coming in 6.9, and why I tell clients to build on the core API instead of bolting on a beta library that breaks every update.
A client’s store had every core block loaded and a sluggish backend to show for it. Here is how we worked out which WordPress core blocks to unregister, and the snippet we used.
A client wanted their marketing team to pick which custom product attribute shows on a page, from the editor. Custom block bindings in WordPress 6.9 make that possible, and here is the source registration I used.
How WordPress 6.9’s unique IDs stop Interactivity API directives from clashing on the same element, plus the data-wp-ignore deprecation and new TypeScript helper types.
WordPress 6.9 fixes the parts of the Interactivity API that made SPA-like navigation painful: dynamic scripts and stylesheets load on their own, nested router regions update, and a new attachTo property can place a region anywhere.
A tiny custom post type tweak turned into a two-month project because we chased perfection instead of shipping. Why incremental development, the way WordPress 6.9 itself is built, works better.
WordPress 6.9 strips out the old Internet Explorer compatibility code. If you run an older custom theme, here is what tends to break and how to audit and modernize the CSS and JavaScript behind it.
WordPress is moving the block editor into an iframe in 6.9 and 7.0. Blocks on apiVersion 2 or lower throw warnings now and can break later. Updating to apiVersion 3 is the fix.