Claude Code routines promise mildly clever cron jobs
Plus Anthropic has redesigned its Claude app
Anthropic has made it easier to automate Claude-oriented tasks without relying on autonomous agent software.
The AI biz on Tuesday introduced a cloud service called routines that allows customers to run Claude Code automations on the company's infrastructure, which hasn't been all that reliable lately.
"A routine is a saved Claude Code configuration: a prompt, one or more repositories, and a set of connectors, packaged once and run automatically," the company explains in its documentation. "Routines execute on Anthropic-managed cloud infrastructure, so they keep working when your laptop is closed."
Routines are somewhat similar to other forms of scheduled tasks, such as cron jobs, GitHub Actions, or AI agents, but not entirely.
Cron jobs and GitHub Actions run set scripts at set times or following specified events, generally without dynamic input from an AI model.
Claude Code routines prompt an AI model on a schedule or a pre-defined trigger or webhook and can take different actions depending on the context they encounter and the connectors that have been made available.
An AI agent would be an ongoing process that maintains state and involves AI model interactions with various tools and data sources.
So a routine could be thought of as a dynamic cron job or a trigger-driven, short-lived agent.
- Claude is getting worse, according to Claude
- Microsoft's massive Patch Tuesday: It's raining bugs
- Nvidia slaps forehead: I know what quantum is missing - it's AI!
- Man suspected of Molotov attack on Sam Altman's home charged with attempted murder
Anthropic suggests routines might be useful for tasks like verifying software deployment – the model scans CI/CD output, checks for errors, and posts a report – or triaging alert messages.
The service is available to Claude Code users on Pro, Max, Team, and Enterprise plans, so long as they have Claude Code on the web enabled. Routines usage applies to subscription usage limits, and there are also daily limits – Pro users can run five routines per day, Max users get 15, and Team/Enterprise users get 25. Usage beyond that can be billed as metered overage if extra usage is enabled.
Also on Tuesday, Anthropic announced a revision of its Claude Code desktop app. It's still based on the Electron framework, which is not beloved by fans of artful native code for its size and inefficiency. Then again, LLM-generated code is not believed by fans of artful native code either, to say nothing about the huge amounts of memory and storage required to make AI work.
"The redesign brings more commonly-used tools into the app, so you can review, tweak, and ship Claude's work without bouncing to your editor," the company explains in its blog post.
It includes an integrated terminal, an in-app file editor, a faster diff viewer, and an expanded preview area.
The salient detail here is "without bouncing to your editor" – Anthropic wants to own the interface through which developers interact with Claude. It would rather not have customers access its AI service through a VS Code plugin or third-party harness like OpenCode (already excommunicated from subsidized subscription usage).
Anthropic also touts its app's ability to manage multiple sessions, an effort, the company insists, to capture how developers actually work with AI models: "kicking off a refactor in one repo, a bug fix in another, and a test-writing pass in a third, checking on each as results come in, steering when something drifts, and reviewing diffs before you ship."
By the way, that sort of multitasking burns through tokens far more rapidly than judiciously applied, carefully reviewed AI assistance. ®
