When rewriting WordPress core in C/C++, I chose a different architecture, which by design blocks a bunch of common exploitation attempts.
Automated scripts/scans constantly test for common setup mistakes on popular frameworks like WordPress.
LLMs can be trained to perfom deliberate exploitive behaviors, just like a computer program.
Just mark my works for when your LLM + harness combo silently uploads your data to a foreign entity and conceals its doings.
Jonathan Hale retweeted
Replying to @Denche2005
php is the kitchen that cooks food for a lot of websites.
someone found that if you shout a large order, the cook panics.
then stranger can walk in and put something rotten on plate waiting to be served to a customer.
The PHP situation is super unfun.
Not saying BookPress is generally safer than #WordPress at the moment, but at least it doesn't hat that bug.
Single command migrate, since command back.
I feel like I developed quite an acuity for AI slop.
Maybe the same verbal fingerprint thing that anthropic is using to fingerprint AI text trained by reading tons of AI responses.
It's so easy to recognize AI writing now, even where people try to conceal it.
Not every product needs to integrate its own agent/harness. Some products are better off just implementing an interface (CLI or MCP).
I just want to connect my agent to your product, not be forced to use whatever you think is the best LLM / agent. Yours does not have my local memory.
Do I care too much about the wording of sales e-mails?
I find it extremely hard to use AI for sales, because of the amount of crappy AI slop copy I find in my inbox every day.
Does it even matter?
With open weights models trailing the frontier I always wondered if we wont at some point get to "good enough", where the frontier might be better, but the job can be done just as well with the last generation.
With Opus 4.8 I felt like the "unlock" for AI being usable for me happened, and anything newer is just slightly better.





