In partnership with
Hey everyone,
Every developer knows this exact feeling.
You give Claude Code a task, step away to grab coffee, come back ten minutes later and it is just sitting there. Frozen. Waiting for you to click approve on something completely harmless.
Or you turned off all the safety checks to stop the interruptions. And came back to a broken codebase.
Both options were bad. Until yesterday.
On March 24, 2026, Anthropic launched auto mode for Claude Code. And it is exactly the middle ground the industry needed.
Primary Sponsor
The Key to This $240B Market Is in Your Bloodstream
Every year, $240B is spent on treating the symptoms of osteoarthritis. But not a single therapy has been able to actually stop it. The answer, it turns out, has been inside us all along.
A startup named Cytonics discovered the human body already produces a protein designed to protect cartilage. It just doesn’t produce enough where it's needed most. So Cytonics harnessed it.
Their first-generation therapy has already treated 10,000+ patients. Now they've engineered a 200% more potent, mass-producible version pushing toward FDA approval.
If approved, it could be the first therapy to actually halt cartilage destruction and promote regrowth in a market that has never had a real solution. Claim a piece at the pre-clinical stage as an early-stage investor before March 26 to receive time-sensitive investor bonuses.
This is a paid advertisement for Cytonics Regulation CF offering. Please read the offering circular at https://cytonics.com/
BACK TO POST
The problem it solves
Claude Code is not just a code suggester. It can run shell commands, create directories, move files, commit to Git, and run tests. That power is why developers use it. And exactly why they worry about it.
Before auto mode, you had two choices.
The default mode meant approving every single file write and bash command. Safe but completely impractical. Research from UC Irvine found that knowledge workers take more than 20 minutes to regain full focus after a single interruption. Constant approval prompts were not just annoying. They were destroying productivity at a measurable level.
Skip permissions mode removed all safety checks. Claude ran free. Developers went to lunch. Came back to weeks of cleanup.
How auto mode actually works
Auto mode sits in the middle. Before every tool call, a built-in classifier reviews the action and specifically looks for three things.
Mass file deletions
Sensitive data should not be sent anywhere
Malicious code execution
If the action passes the check, Claude proceeds on its own. If it fails, the action is blocked, and Claude finds a different approach. Only if Claude keeps pushing against the boundary does it escalate to a human prompt.
It also watches for prompt injection attacks, where hidden malicious instructions inside the content Claude is processing could trick it into doing something the developer never intended.
The result is a multi-hour refactor that runs clean. Files move, imports update, types regenerate. But if Claude tries a wildcard delete that sweeps too broadly, blocked. If it tries to read environment files or push logs containing secrets somewhere, blocked.
No silent disasters. No broken repos.
The honest catch
Anthropic is not overselling this. The classifier can still miss risky actions when context is unclear, and it occasionally flags harmless commands too. Auto mode only works with Claude Sonnet 4.6 and Opus 4.6 right now. It may slightly increase token costs on high-volume operations. And Anthropic strongly recommends sandboxed environments only, not production systems.
How to turn it on
CLI: run claude --enable-auto-mode then Shift+Tab to activate it. VS Code: Claude Code settings, permission mode dropdown. Desktop: Organization Settings then Claude Code. Currently available to Team plan users as a research preview.
Secondary Sponsor
What 200K+ Engineers Read to Stay Ahead
Your GitHub stars won't save you if you're behind on tech trends.
That's why over 200K engineers read The Code to spot what's coming next.
Get curated tech news, tools, and insights twice a week
Learn about emerging trends you can leverage at work in just 5 mins a day
Become the engineer who always knows what's next
LET'S FINISH
Why this matters
Claude Code just crossed $2.5 billion in annualized revenue, up from $1 billion ten weeks ago. Auto mode is part of a bigger push from Anthropic to build a full autonomous engineering platform, not just a chat tool with coding skills.
The real question this raises is not technical though.
How much do we trust AI to decide on its own what it is allowed to do?
Today the answer is the safe stuff only, with a human always one step away.
But that line is moving fast.
Catch you next time,
— RAJA TAHOOR AHMAD
P.S. If you manage a dev team, forward this. Auto mode is worth testing this week.
