Dans l'interview du jour, @osteel.bsky.social partage sa vision d’un code plus maintenable face à l’IA, sans perdre le contrôle du pipeline. On vous laisse découvrir cette interview enrichissante, avant d'en apprendre plus ensemble lors de son talk le 22 mai à l'AFUP Day 2026 Lille !
buff.ly/o7t3Y7v
Our Executive Director, Elizabeth Barron, has been listening to your feedback about the PHP ecosystem to help inform The PHP Foundation's 2026-2027 strategy. She shares what she's learned in our latest blog post. Exciting opportunities await! ✨ 🚀 thephp.foundation/blog/2026/04...
#php #phpcommunity
My Laravel Live UK speaker card
Guess who’s speaking at Laravel Live UK in June 😌 https://laravellive.uk
I enjoyed this post about how engineering roles are being redefined in the age of AI. What are the traits that truly matter in employees and candidates? In short: understanding > implementation
// For users of Laravel Boost php artisan boost:add-skill peterfox/agent-skills --skill composer-upgrade // For everyone else npx skills add peterfox/agent-skills --skill composer-upgrade
Just used Peter Fox's `composer-upgrade` skill on PixelWatcher. Amazing work. It breezed through minor dependency upgrades, reported on the only major one, pulled the corresponding changelog, and checked whether my code would be impacted. Then it ran the test suite. Check it out!
Lille 😌
I have to say, having Claude use Nightwatch's MCP server to pull the latest issue and suggest a solution within minutes is freaking cool
3️⃣ Add the middleware to a new group that you assign to the route file
4️⃣ Create caching rules for your domain in Cloudflare:
- One that bypasses stateful routes
- One that bypasses session cookies
- One that targets static pages specifically
How to cache static pages in a Laravel application (full post at the end of the thread):
1️⃣ Create a separate route file for static pages
2️⃣ Create a dedicated middleware adding cache headers to responses
I see. I guess the difference is whether there’s a human in the loop or not. Fully autonomous, or someone reviewing and approving at some point. At the moment, in most cases, I don’t think full autonomy is a good idea!
As in "They are replacing the CRAFT and THINKING"
I guess that’s why so many people say that "this time it’s different".
I’d push back on the thinking though – if there’s one thing that hasn’t been replaced yet (at least not everywhere), it’s decision and judgement.
ChatGPT is far too verbose, and no amount of customisation seems to fix it. I end up losing my train of thought in most conversations.
We're expected to manage these tools' context windows, but what about ours 😩
Amazing post by Linear acknowledging their original product is doomed. I just got comfortable keeping all useful context inside PixelWatcher's repo, but as I get involved with more early-stage products, a context-management tool starts to make sense
Yes I think you might be right. I need to tailor it so it triggers when stopping in the middle of that particular flow somehow, but this looks like a good lead. Thanks!
Think so! But I will double check tomorrow. In my case it’s basically a skill calling other skills, but maybe there’s a better way to do it
Yep. The flow is composed of various tasks (plan, implement, review, etc.) that are supposed to be performed in sequence by the main agent, which gives each task to a separate subagent. But the main agent will sometimes stop after a subagent is done, instead of moving on to the next task
It’s not stopping because it needs some permission – each step is completed by a subagent, but it will sometimes stop after a subagent is done, instead of moving on to the next step. And if I ask "why did you stop?", it will "oh yes, sorry" and then resume the flow 🤷🏻♂️
Did my first attempt at automating an entire workflow with Claude today (plan > implement > simplify > review > test > PR), the idea being to get to a decent first draft for a task, which I can review and refactor. But it keeps stopping at random steps. Not sure how to fix this
Not exactly kind on the retina, either
Screenshot of the OP's status line in Claude Code. It shows: * Current directory * Current Git branch * Current model * Context usage * Session usage * When it resets * VIM mode * Model mode * Open PR
Finally taking the time to use my faster.dev subscription and look at the video series. I almost skipped the "Claude Code from Scratch" one entirely, but decided to have a quick look anyway. Guess what – I didn't know you could customise the status line 🤦♂️
I've now moved all my skills to a dedicated repository, with a small script to install them for Claude and Codex. I guess the upside of all this is that I can now switch from Claude Code to Codex CLI with little friction, and that's definitely a win
github.com/osteel/agent...
Kind of means that at least you’re still needed though (#glassHalfFull)
Which is also why I'm building a SaaS – to prove (to myself, first and foremost!) that I can handle it, and to come up with a repeatable workflow and AI setup. Going well so far, but yes, I'm going to have to do client work at some point in the near future :)
I took some time off at the end of last year (paternity leave), and now I'm trying to transition from "backend contractor" to "MVP builder" (mostly because the contracting market's been awful for the past 3 years, and there's no recovery in sight)
Yeah I think at this point everyone's aware that these tools are currently massively subsidised. $200 per month is actually a very good deal, in the grand scheme of things!
I guess it's also the way they did it sneakily by hiding it behind a "2x usage promotion", then denied it happened, then finally admitted it. On a more personal note, I'm not doing any client work atm, so $200 per month is a bit steep. I'll happily pay for it when I'm actually getting paid though
Overall, Codex CLI feels much slower and clunkier than Claude Code. As I can't use Codex because of my Docker-based setup, currently, the overall experience is far better with Claude Code (when it works)
