In partnership with
Your Claude got an upgrade last week. You probably didn't notice.
On June 30, Anthropic made Sonnet 5 the default model on Free and Pro.
So the Claude you opened this morning is smarter than the one you used last month.
What actually changed, and the one shift that gets you the benefit.
Now a word from today's sponsor:
200+ Claude Prompts Top Professionals Actually Use at Work
Claude can be your analyst, editor, and strategist.
But most professionals are using it to fix grammar.
These 200+ Claude prompts take it from grammar tool to your most powerful AI work assistant.
Sign up for Superhuman AI and get:
200+ ready-to-use Claude prompts to get real work done in minutes — researched, tested, and used by professionals at Google, Microsoft, and NASA
Superhuman AI newsletter (4 min daily) so you keep learning new AI tools and skills to stay ahead in your career — the prompts are just the beginning
Okay, what's different.
Sonnet 5 is built to finish jobs, not just answer questions.
It plans across steps. It uses tools. It checks its own work before handing it back. And it reasons noticeably better on hard, multi-part problems.
In plain terms: it can take on bigger, messier tasks and actually carry them to the end.
But there's a catch.
Most people are still using it like it's 2024.
They ask one small thing. They get one small answer. They never see the upgrade.
The shift is this. Give it the whole job, with context, in one go.
Instead of a one-line request, hand it the goal, the background, and what "done" looks like.
Try this the next time you'd normally fire off a quick ask:
Goal: [what you actually want, the finished outcome].
Context: [everything Claude needs, paste it all, don't hold back].
What "done" looks like: [the format, the length, the standard].
Do the whole thing. Then check your own work and flag anything you're unsure about or any assumption you made.That last line matters.
Sonnet 5 will genuinely review its own output and tell you where it's shaky. Older models mostly didn't.
Two more things worth knowing.
One, it has a huge memory for a single conversation. So paste the whole document, the full email thread, all the notes. More context in, better work out.
Two, it thinks on its own when a task is hard, and answers fast when it's easy. You don't have to manage that. Just give it the room.
The takeaway is small, but it changes everything.
Stop treating Claude like a search box. Start treating it like a capable assistant you're handing a real assignment.
That mindset, plus the setups that support it, is the whole of Claude Mastery.
The full system for getting a team's worth of work out of Claude. Step by step, no tech background needed.
Quick one.
What's the biggest job you'd hand Claude if you trusted it to finish? Hit reply.
Talk soon,
Zephyr
