How about this Claude Code @claudeai statusline, so you don't have to keep a browser window open and constantly check if you'll be able to finish the task before the session reaches its quota limit and restarts?
Did you like it? This is a new project, just access my repo on Github, give it some love, click the star, and download it, follow the README and make Claude's statsline look great.
https://t.co/iMQ9wVuh0d
It shows:
- current directory
- branch
- files and lines changed (quantity)
- Thinking time
- Total tokens up and down
- Total cost of the session
- 5-hour session: time to restart and percentage used
- Weekly session: time to restart and percentage per model or extra usage
I love that non-deterministic AI is becoming the glue in between human thought/language and reliable, deterministic systems like @laravelshift
And I think this is going to become increasingly important as compute subsidies subside and effective token costs increase
For our NativePHP Desktop users
Our PHP binaries have finally been updated to the latest versions and the workflows fixed
We even managed to squeeze in PHP 8.5 support! โ๐ผ
All the details here:
https://t.co/Iex5qKUE0v
Introducing Claude Opus 4.7, our most capable Opus model yet.
It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back.
You can hand off your hardest work with less supervision.
So excited to share my first plugin on the @nativephp marketplace! ๐
โnativephp-in-app-reviews lets you trigger native App Store and Google Play rating prompts without breaking the user experience.
โกโ InAppReviews::requestReview();
https://t.co/69B8xNJCQk
It has begun
Our first batch of premium *community* plugins ๐
3 awesome plugins by core contributor, @SRWieZ:
Contacts
Calendar
Screenshots
All 3 available for iOS and Android
๐ My first ๐ก๐ฎ๐๐ถ๐๐ฒ๐ฃ๐๐ฃ ๐ฝ๐น๐๐ด๐ถ๐ป is live on the official site!
Introducing ๐๐ต๐ฒ๐๐ป๐๐ถ๐ป๐ฑ๐ณ๐ฟ๐ผ๐ป๐/๐ป๐ฎ๐๐ถ๐๐ฒ๐ฝ๐ต๐ฝ-๐ฎ๐๐ฑ๐ถ๐ผ bringing seamless audio features to NativePHP projects.
Check it out: https://t.co/jPvnpX0Inq
#PHP#OpenSource Thanks to @nativephp
We just released Cloud CLI v0.2.0, which introduces... ๐ฅ
๐๐๐๐๐ ๐๐๐๐๐๐
You can now tinker in your Laravel Cloud environment right from your terminal.
Bonus: If you want to tinker from your IDE, just use the --๐๐๐๐๐๐ flag.
Happy tinkering ๐งช
NEW FIRST PARTY PLUGIN ๐ฅ
Local notifications. Fully native. Powered by PHP.
No Swift. No Kotlin. No JS bridges.
Just shipped a new NativePHP plugin ๐ฅ
ULTRA subscribers get it for free, link below ๐
I got this view shared at weekend in one of my posts, just loved the perfect description of the real problem of non technical vebe coders: "They never learned the unsexy parts."
What this means?
They never learned: exceptions handling, log tracking, ideal relational modeling, record or file pruging, disk space handling, api services migration, legacy maintenance (yest with luck if the app do not ruins in 6 months, it's already a legacy to be maintained while new features need to be added) .. so .. as I see .. the future is foggy for them
With NativePHP, you never need to open Android Studio ๐ ๐ผ
With all the time and money you'll save, you can go buy yourself a *real* grill!
Thank us later
We want the #PHP language and community to grow, so we HAVE TO embrace PHP being able to do *more* than the web or CLI
That means supporting true async, native abstractions and more
The future of language choice is going to be based on:
- capability
- stability
- community
People doing "vibe coding" are going to be lost when simple tasks ... like switching disk storage from local to AWS S3 or DigitalOcean Spaces ... become necessary because a VPS disk is full. They likely didn't even consider document purging while "vibe coding" the software. This is just one simple oversight AI may not handle given the "poor prompts" used by non-technical people.
I'm not even focusing on security, which is the first issue developers point out regarding vibe coders. In my view, while AI tool developers tend to address the loudest market complaints within their workflows .. like security ... the tiny, hidden flows that we as software engineers carefully build into solutions will start breaking these vibe-coded apps once they gain clients and begin to grow.
I tested Antigravity on the holiday, with Gemmini 3.1 Pro (High)... conclusion... no thanks.
Cursor still performs better if you're thinking about an AI IDE.
Claudinho remains the best in my context.
adversarial-review is the way to apply a different model to the one that planned / coded .. I've being using coderabbit already and it always get something the own claude code reviewer couldn't see, because it's like the developer to test it's own code, you don't want to break it ;-)
Starting today you can use Codex in Claude Code ๐
/plugin marketplace add openai/codex-plugin-cc
Try it out today with:
/codex:review for a normal read-only Codex review
/codex:adversarial-review for a steerable challenge review
/codex:rescue to let codex rescue your code
Enjoy Codex-ing!