A simple question: "Why is my plugin not first in the result, when I search for it's exact name?" lead to a series of interesting discoveries about the free plugins directory.
I've seen this question asked and answered many, many times. And it seems that the answers depend on who you ask:
- WP Users: "I have no idea... it just is."
- Non-WP Users: "Because it's old/PHP. Use Next/Nuxt/Astro/Webflow/Wix/Ghost/...."
- Theme developers:
"It's plugins or hosting. Or other themes - ours is fast."
- Plugin developers:
"It's theme or hosting. Or other plugins - ours is fast."
- Hosting company: "It's plugins or theme. Or other hostings - come to us we're fast."
- WordPress "Experts": "WordPress powers 43% of the web(...)"
- WordPress Core Performance: "We'd like to make it fast, but we can't make any drastic changes, because someone should be able to run their blog on a solar-powered potato..."
- WordPress Performance Experts: "It's complicated... I can make an audit, it will take around 10h and cost $999"
- WordPress Cache Plugin developers: "It's not if you cache everything!"
- WordPress Performance Plugin developers: "There are many reasons, but our plugin fixes all of them!
*unless it doesn't - then you need to switch hosting"
( yes, it's a satire )
I build this tool for myself ( I'm running many services ) - but then one of the clients suggested white-labeling it and offering as a product.
I did that and I think it looks quite cool.
It's lifetime purchase so I didn't know how to price it.
I think it's worth $99 ( it saves 2-3 hours of doing it with AI ) - but maybe I should make it dirt cheap and have higher scale instead?
What do you think?
These are only free plugins, that were submitted to wp.org - not counting all of the others that have not been submitted.
My main concerns about that ( in no particular order ):
On one side - I understand why that happens. The cost of creating own plugin has dropped probably around 10x now.
But I'm really afraid that this will seriously hurt WordPress as a whole on the long run.
Someone published a piece claiming a theme author was inflating his active install counts on WordPress.org. I read it, got annoyed at how plausible it sounded, and decided to check it myself instead of taking anyone’s word for it.
That took a day. What I found is worse than the original accusation, and it has nothing to do with that one author.
The number in the plugin directory that says “10,000+ active installations” is the softest number in the whole ecosystem. It has no public history, it’s rounded to one significant digit, it costs about a third of a kilobyte to fake, and it drives your position in search. Every other number on that page can be checked by anyone. This one can’t be checked by anybody outside wp org, including the plugin author whose reputation is attached to it.
If your WordPress site runs Redis ( it should ) and the status screen says Connected, you know one thing - a socket opened. That's it.
That test cannot fail in most of the ways an object cache actually fails in production. I deliberately misconfigured a cache on a test site and filled it, and the site started serving HTTP 500 to every visitor while WordPress carried on reporting that an external object cache was active.
Redis is great, but it's not "fire and forget". Post covers some of the pitfalls and ways to avoid them.
I built a small benchmark to answer one question: is Claude Opus 4.6 actually worse than 4.8, or does it just feel older?
It scores the half of code review that most evals ignore, which is restraint - not flagging correct code that looks suspicious.
Four models, four task families, one-shot and mechanically scored. The two Opus models came out statistically tied, and they cost the same, so there's no reason to pick one on price. Across a real tier gap, Sonnet 5 matched Opus on almost everything while running fastest and costing a third as much. The surprise was Haiku 4.5: it was both the least accurate model and the slowest, because adaptive thinking makes it burn a huge token budget on simple tasks.
My usage over last week by model
My point is: you don't really NEED Fable 5 to ship things
Hot take: Vast majority of the improvements in working with LLMs in general and Claude Code in particular - is in harness/skills - not that the models got inherently better.
As an example - you can ask Haiku/Sonnet to 'center the div' and they would do exactly the same good job ( possibly faster)
There was no announcement - but I just noticed this a moment ago. I'm sure it wasn't like this a few hours earlier.
TL;DR
AskUserQuestion dialog now auto-answers after 60 seconds (with a 20-second on-screen countdown) and tells the model to proceed on its best judgment.CLAUDE_AFK_TIMEOUT_MS and CLAUDE_AFK_COUNTDOWN_MS. Set the timeout to 2147483647 and it effectively never fires.strings over the local binary and bisecting four versions. Every command in this post is reproducible on your own machine.I'm supporting many client sites - both retainers and one-off optimizations. More often than not, while looking for the sources of the slowness I end up finding an enabled debug.log that takes 5G+ and is full of notices and deprecation messages.
So on one side it's good as this is the mine of useful information about website.
On the other side - sifting through the 5G of crap takes ages.
Do you have any good solutions to deal with that?
From 19 June 2026, every shop selling to EU consumers has to put a clearly labelled, one-click "withdraw from contract" function in the storefront. That's Directive (EU) 2023/2673, which adds a new Article 11a to the old Consumer Rights Directive.
WooCommerce has no native support for it - there's just an open request (#65443) sitting there with no roadmap.
I've tested 11 plugins that do that and selected the best one ( in my opinion )
Over the past few months I've benchmarked the performance of ALL 60k+ plugins from WordPress Plugins repository and graded them.
Turned out that most of them had negligible impact on performance. Then one of the commenters asked: "So, did you install all 50,106 top performing plugins and see what happens?"
(source: https://www.reddit.com/r/Wordpress/comments/1u00akj/comment/oqhot8e/ )
I turned it into a challange: "How many "fast" plugins can I really run on a single WP instance at the same time?"
So I spin a Docker environment locally, took the list and started installing...
Here's the full article:
https://makewpfast.com/how-many-plugins-can-wordpress-handle/
TL;DR: Even if the plugin is 'small' or 'fast' it DOES have a tiny footprint, and this compounds.
I stopped after installing 223 plugins (without touching the limit)
TL;DR; It's WAY more about QUALITY than it is about QUANTITY.
A major release of the WP Multitool (1.3.0) that massively improves the Image Optimizer module - adding the advanced detection of the unused images and thumbnails and allowing to remove them.
WordPress still registers blocks even if you're not using them, and not only basic WP, WooCommerce too. And even if it doesn't take MUCH time - it still takes some time on every request.
So I've made a free plugin that just... removes them https://wordpress.org/plugins/zero-blocks-given/