Anthropic launches a repeatable routines feature for Claude Code as a research preview, allowing developers to schedule and automate software development tasks
Anthropic's Claude Code has a new repeatable routines feature that works even when your Mac is offline.
9to5MacZac Hall
Context & Ripple Effects
Anthropic has been extending Claude Code beyond an interactive coding assistant: web and iOS access arrived in beta, then Cowork was introduced for minimally prompted task automation, followed by computer-use capabilities in the desktop app.
Routines add recurrence to that progression. The important change is not a new model capability alone, but a mechanism for turning one-off developer interactions into repeatable work on a Mac.
First-order effects
Claude Code users in the research preview can schedule recurring software-development tasks rather than manually initiating each run; the feature can operate while the Mac is offline.
Anthropic further ties Claude Code’s value to its desktop workflow, alongside the recently previewed computer-use capability for Pro and Max subscribers.
Second-order effects
Competing coding assistants face pressure to support reliable recurring workflows, not just prompt-driven code generation, if developers begin treating agents as background task runners.
For developer teams, routine automation can shift evaluation toward setup, oversight, and repeatability of workflows; offline execution makes local-machine integration a more consequential product consideration.
Third-order effects
If these previews mature, AI coding products may evolve from chat-based helpers into workflow-native execution surfaces that coordinate recurring development work across desktop tools.
That shift would increase the importance of controls around what automated routines can access and execute, especially as Anthropic is also pursuing tougher AI-safety rules at the state level.
The trend: Developer AI is moving from on-demand assistance toward persistent, workflow-native agents that can carry out recurring work with less manual initiation.
Now in research preview: routines in Claude Code. Configure a routine once (a prompt, a repo, and your connectors), and it can run on a schedule, from an API call, or in response to an event. Routines run on our web infrastructure, so you don't have to keep your laptop open. [ima…
Claude Code Routines are here! In addition to a schedule, you can now trigger templated agents via GitHub event or API - with our infra & your MCP+repos They've changed how we do docs, backlog maintenance and more internally at Anthropic Get started at https://claude.ai/... [vide…
Been incredible watching this labor of love from a incredibly talented, kind, and thoughtful human being. He's channeled weeks of feedback into an amazing redesign of the Claude Desktop app!
The redesign also adds an integrated terminal, file editing, HTML and PDF preview, and a faster diff viewer, all in a drag-and-drop layout you can arrange to your preference. Your CLI plugins work exactly as they do on the command line.
We've redesigned Claude Code on desktop. You can now run multiple Claude sessions side by side from one window, with a new sidebar to manage them all. [video]
Today we're launching a rebuilt version of Claude Code on desktop. The app has been redesigned for the ground up to make it easier than ever to parallelize work with Claude. I haven't opened an IDE or terminal in weeks. Excited for you all to give it a shot! [video]
claude now runs autonomously with your computer turned off. 24/7 coding agent. this replaces $10,000's of engineering time and competes directly with Pager Duty's $500M val it works while you sleep. reviews code, builds new features - ready for you to click publish when you [imag…
there's really no point in building a general use agent bc if you get a little bit of motion the frontier labs are gonna come for your money... brutal tbh