Honest take on opus 4.8, not impressed (yet)
Does more testing on its own, which is good, but also time consuming and sometimes not needed
Overcomplicates- had a mobile issue and it tried to create a new JS function when the simple fix was CSS
So for now more hand holding...
Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Available today at the same price.
Claude Code 2.1.139 has been released.
50 CLI changes, 1 system prompt change
Highlights:
β’ Added /goal command to run tasks across turns until a set completion condition, with live elapsed/turns/tokens
β’ System prompt compaction is now silent (no trimming alert), which may obscure that prompts were trimmed
β’ Compaction prompt asks model to preserve sensitive instructions, reducing loss of user intent during trimming
Full details are in thread β
Agents make ugly UIs because they've never seen good design.
We've been fixing that, 2,000 DESIGN.md files from the world's best products, structured for a model to read and learn. Colors, type, spacing, layouts and more.
Free. https://t.co/mJaKNHba0O
Claude Security is now in public beta for Claude Enterprise customers.
Claude scans your codebase for vulnerabilities, validates each finding to cut false positives, and suggests patches you can review and approve.
Question for #Drupal devs:
I'm building an AI chat app that creates fully custom Drupal sites from prompts. Think ChatGPT but the output is a working site in ddev.
Useful, or more hassle than just prompting Claude/GPT while you code it yourself?
Over the past month, some of you reported Claude Code's quality had slipped. We investigated, and published a post-mortem on the three issues we found.
All are fixed in v2.1.116+ and weβve reset usage limits for all subscribers.
Never got to try out Loveable due to always paying for Claude, so I'm excited to give this a try and see how it compares against regular Claude generated UI/UX!
Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude.
Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on the Pro, Max, Team, and Enterprise plans, rolling out throughout the day.
Building a tool that generates full Drupal sites from chat. Describe it, approve the design, and get a real DDEV environment you can log into.
No more starting every project from an empty admin screen.
Mythos was able to break out of a sandbox and send an email π€―
This is huge! How can we keep them wrangled or should they just be out in the wild roaming like animals?
If a sandbox can't keep it contained then are AI-first & only operating systems next?
Introducing Project Glasswing: an urgent initiative to help secure the worldβs most critical software.
Itβs powered by our newest frontier model, Claude Mythos Preview, which can find software vulnerabilities better than all but the most skilled humans.
https://t.co/NQ7IfEtYk7
But anyway here's the TL;DR π
I finally got https://t.co/dQUT3nlECN to create a customized Drupal website from just one chat dialog.
Which helped me finally update it since 2020 π
https://t.co/ObYZaUAXWP
When you have that ah-ah moment and π₯
I was hammering away expecting different results. After manually building along with Claude, I could then copy that proven workflow.
With that workflow I was able to fit in the missing pieces I didn't know I was missing.
Along the way I found other gaps and fit them into the workflow until a polished(ish) version could finally produce the customized Drupal site.
Still not perfect but doing tasks manually first to see the ins and outs can come in handy with finding pieces you don't know about.