Claude Code Gets a Mission Control and Routines That Run While You Sleep
On April 14, Anthropic dropped two things at once: a complete redesign of the Claude Code desktop app and a new feature called Routines in...
In this article
Mission Control: Managing Agents Instead of Writing Code
The biggest visual change in the redesigned app is the new sidebar — Anthropic calls it 'Mission Control.' It gives you a single view of every active and recent Claude Code session, filterable by status, project, or environment.
This is a philosophical shift in how Anthropic thinks about what developers actually do. You're not typing prompts and waiting for an answer. You're running a refactor in one repo, fixing bugs in another, writing tests in a third — and you need to monitor and steer all of them without losing track. Mission Control is the answer to that.
The sidebar also surfaces diffs before they ship, which addresses one of the biggest enterprise concerns about agentic coding: you want to review what Claude did before it actually lands in production. That's now a first-class part of the interface instead of something you have to go hunting for.
Routines: Three Types, Very Different Use Cases
Routines are the more technically significant announcement. The core idea: instead of agent work being tied to your local machine (laptop open, session running), Routines move execution to Anthropic's web infrastructure. A task that needs to run at 2 AM can run at 2 AM without your laptop being on.
Three types of Routines launched in research preview:
- Scheduled Routines — Cron-style automation. Runs Claude on a schedule for repeating tasks like nightly bug triage from a Linear backlog, docs-drift scanning, or dependency updates. Set it once, review results in the morning.
- API Routines — Dedicated HTTP endpoints and auth tokens so external systems can trigger Claude. Your Datadog alert fires, it calls the API Routine endpoint, Claude investigates and files an issue. Your CI/CD pipeline fails, Claude gets triggered automatically to diagnose it.
- Webhook Routines — Currently GitHub-focused. Claude listens for repository events — PR comments, CI failures — and automatically opens a session to respond. The reviewer leaves a comment asking for a refactor; Claude starts the work.
Scheduled Routines are roughly equivalent to 'cron for Claude.' API Routines are 'Claude as a service your other tools can call.' Webhook Routines are 'Claude that reacts to your repo in real time.' Together, they make Claude Code a 24/7 autonomous worker, not a session-based assistant.
Also On April 15: Claude Had an Outage
Separately — and somewhat ironically given the launch — on April 15 Claude experienced elevated error rates across the chatbot, API, and Claude Code. The outage started around 10:53 AM ET, peaked with ~6,000 users reporting issues on Downdetector by 10:42 AM ET, and was resolved by 1:50 PM ET. About 3 hours total.
Anthropic's status page handled it cleanly, with updates every 15-30 minutes. Not their first outage, but notable because it happened the day after the Routines launch — which depend entirely on Anthropic's infrastructure being available. If a Scheduled Routine is set for 2 AM and Claude's API is down, the Routine silently fails.
Worth keeping in mind as you evaluate Routines for production use: they're only as reliable as Anthropic's uptime. Right now that's very good, but it's not SLA-backed in the same way enterprise infrastructure typically is.
What This Means For You
If you're a solo developer: Scheduled Routines are the immediately useful thing. Set a nightly code review, a weekly dependency check, or a PR summary to run on schedule. These are real hours back per week.
If you're on an engineering team: API Routines are where the real leverage is. Wiring Claude into your existing alerting and CI/CD means you can get a first-pass diagnosis on every failure automatically — before anyone opens their laptop. That's the kind of change that shifts the on-call experience.
The Routines feature is in research preview, which means it'll change. But the direction is clear: Anthropic wants Claude Code to be something that's running for you even when you're not at your desk. That's a fundamentally different product from the coding assistant it was six months ago.
Sources & Further Reading
• We tested Anthropic's redesigned Claude Code desktop app and Routines — VentureBeat, April 15, 2026 • Anthropic outage: Elevated errors across chatbot, Claude Code and API — CNBC, April 15, 2026
Get the Claude playbook in your inbox.
One weekly email for Claude and Claude Code users. Real workflows, no hype. Subscribe and we send you The Claude Power-User Cheatsheet.
— ¶ —
Luke Thompson
Luke Thompson is the founder of The Operations Guide, LLC and editor of The Claude Insider. Based in Jonesborough, Tennessee, he has spent years building AI-augmented business systems and automation workflows for operators and teams. He began working with large language models in production well before the current wave of consumer AI tools, integrating them into client workflows, content pipelines, and operational infrastructure. At The Claude Insider, he writes about Claude with the specificity of someone who uses it daily as a professional tool — not as a reviewer or commentator, but as a builder. His coverage focuses on what actually works: prompt patterns, API integration strategies, agentic workflows, and the real-world tradeoffs that practitioners face. He is not affiliated with Anthropic, PBC.
Articles are researched and drafted with AI assistance, reviewed and edited by Luke Thompson.
Know where AI can pay off in your company.
Take the free two-minute AI Readiness Assessment. See your score, the two gaps holding you back, and the next move worth making.
