OneRange

What's New in Claude for Developers

A short refresher on the latest Claude, Claude Code and Claude API updates for developers

What this course covers

The course runs across 7 topics, each one a short adaptive session rather than a recorded lecture. The tutor explains the idea, works an example, checks understanding, and adjusts the next step to the answer given.

  • Sessions now start in auto mode — What's different: New terminal and VS Code sessions now start in auto mode, where a safety check reviews each action, unless you have set a different default mode. Why it matters for your work: Claude asks you to approve far fewer steps, so work moves faster, while risky actions are still stopped or flagged. Try it (about 2 minutes): Start a new session and look at the mode shown by the prompt, then press Shift+Tab to cycle through the modes and back to auto. No access? Here's what you'd see: New terminal and VS Code sessions show auto mode in the prompt, where pressing Shift+Tab cycles through the modes.
  • New API version and major SDK releases — What's different: A new API version (2026-10-21) shipped along with new major versions of the official SDKs and CLI, which use it by default and include some breaking changes. Why it matters for your work: Upgrading an SDK can change how your code behaves, so you need to check the migration guide before you update. Try it (about 2 minutes): Check which SDK version your project uses, then read the migration guide for the 2026-10-21 API version before you upgrade. Needs: Claude API.
  • Fast mode on the older Opus model is being retired — What's different: Fast mode for the previous Opus model was deprecated on October 30 and will stop working on November 30, with no automatic fallback; users must switch to the newest Opus model to keep the speedup. Why it matters for your work: Anyone who pinned the older Opus model with fast mode will start getting errors after the cutoff date. Try it (about 2 minutes): Run /model to see which model your session uses, and if it is the older Opus, switch to the newest Opus before turning fast mode on with /fast. No access? Here's what you'd see: After November 30, the older Opus model's fast mode will fail with errors, requiring users to switch to the newest Opus model.
  • Claude Sonnet 4.5 is deprecated (retires Nov 30) — What's different: Claude Sonnet 4.5 is now deprecated on the API and will be retired on November 30, 2026, with Anthropic recommending the newer Sonnet model instead. Why it matters for your work: Apps that still use this model need to move before the retirement date or they will stop working. Try it (about 2 minutes): Search your code for claude-sonnet-4-5 and test one request with the newest Sonnet model to check the output still fits. Needs: Claude API.
  • A new, faster Sonnet model — What's different: A new Sonnet model was released as a faster, lower-cost partner to the new Opus model. Why it matters for your work: Sonnet is a good pick for quick, everyday tasks when you do not need Opus. Try it (about 2 minutes): Open the model picker, choose Sonnet, ask for a small change, then switch back to Opus if you prefer. No access? Here's what you'd see: The model picker displays a new, faster Sonnet model option.
  • Also new — - A new Opus model is now the default: A new Opus model is now the default, and Pro and Team plans start on Opus. - Block file edits outside your project folder: A new permission setting blocks Claude from editing files outside your project folders. - Guided setup teaches auto mode about your project: A new setup command teaches the auto mode safety check about your project environment. - Long idle chats get summarized automatically: Long idle conversations are now automatically summarized in the background to save time. - Mods, plus a built-in "You should know" helper: Claude Code now supports mods, including a built-in helper that flags potential mistakes. - Claude Code reads AGENTS.md files: Claude Code now reads AGENTS.md files for project instructions when CLAUDE.md is missing. - Review edits one change at a time in VS Code: The VS Code proposed-change view now has accept and reject buttons for individual changes. - Your claude.ai skills and plugins sync to the terminal: Skills and plugins enabled on claude.ai now sync to terminal sessions. - Concise style for shorter replies: A new built-in Concise output style provides shorter replies by skipping step-by-step narration. - Computer use is GA and a browser use tool is new: The computer use tool is now generally available alongside a new browser use tool. - and 16 more: Console Workbench is now Playground; Claude Opus 4.1 retired from the API; /diff now opens a live changes panel; Ask helper agents to work at a set effort level; Get back a prompt you cleared by mistake; Open a terminal session in the desktop app; A new top-tier Fable model; Ultraplan was removed; New Usage and Cost Analytics API; old one is legacy; Claude Tag API for Enterprise admins (beta); The ultrathink keyword now means extra-high effort; Rewind no longer undoes sed edits; Connector permission rules now match exactly; Colored code blocks in VS Code replies; Code Review now has Quick, Standard and Deep levels; Claude API run by Anthropic, reached via Google Cloud
  • Try one on your own work — Pick the one main change from this course that is most useful to your job. Try it in your own Claude on a real task, then paste or describe what you got. The coach gives short, specific feedback. No access to the app or plan it needs? Describe how you would use it on a real task this week. That counts too. There is no score and no pass or fail: finishing this completes the course.

Skills you build

Results are measured against named skills in the OneRange taxonomy of more than 10,000 skills, so a manager sees proficiency per skill rather than a completion tick.

Who it is for

The material is pitched at intermediate level, and takes roughly 20 minutes at a typical pace. Because every session adapts, someone who already knows a topic moves through it quickly instead of sitting through an explanation they do not need.

It sits in the Tool: Claude track of the Vero AI Catalog, and can be assigned to one person, a team, or the whole company.

How it is delivered and assessed

Delivery is conversational and interactive. There is no video to sit through and no slide deck to click past.

Understanding is checked with a lab assessment, with a pass mark of 70%.

Administrators can copy this course into their own library and adapt it — edit the outline, change the duration, swap the assessment format, or ground it in internal documentation so answers cite the company's own source material.

Also in this track

Related

Frequently asked questions

How long does this course take?

It runs roughly 20 minutes at a typical pace. Because every session adapts, someone who already knows a topic moves through it quickly instead of sitting through an explanation they do not need.

How is it assigned to a team?

An administrator assigns it from the Vero dashboard to one person, a team, a department or the whole company, and sees progress and results per person and per skill.

Can we customise it with our own documents and terminology?

Yes. Administrators can copy this course into their own library and adapt it — edit the outline, change the duration, swap the assessment format, or ground it in internal documentation so answers cite the company's own source material.

Course code what-s-new-in-claude-for-developers-1f89e989.