Anthropic launches Routines to run Claude Code autonomously in the background

Claude

Apr 14, 2026 · Updated May 1, 2026

Anthropic introduced Routines for Claude Code, allowing developers to package prompts, repositories, and connectors into repeatable, autonomous workflows. By moving execution to web infrastructure, these agents can now monitor pull requests or fix bugs on a schedule without requiring an active terminal session.

Claude Code, the agentic coding tool from Anthropic, launched a research preview of Routines. Routines allow you to configure a prompt, repository, and connectors once to run repeatedly. Unlike standard terminal sessions, these workflows execute on web infrastructure, removing the need to keep your local machine active or the terminal open.

This update shifts Claude Code from a reactive assistant into a proactive digital teammate. By supporting API endpoints and GitHub webhooks, the agent can now integrate directly into existing developer workflows like on-call triage or security monitoring. It bridges the gap between manual AI chat and fully autonomous, event-driven engineering automation.

You can now schedule Claude Code to attempt bug fixes overnight or trigger it via API to summarize alerts in internal channels. Webhook support currently focuses on GitHub events, such as flagging specific pull requests. Routines are available today for all paid plans with web access enabled.

View the full update on code.claude.com

Claude

@claudeaiApr 14

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. https://t.co/m2XJWYqkf8

72210k

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from Anthropic →

Keep reading

Anthropic enables one-time autonomous background tasks in Claude Code

Anthropic introduced a Once trigger for Claude Code Routines, allowing users to schedule one-time autonomous tasks from the CLI or web interface. This shifts the tool from a recurring automation engine to a deferred execution assistant that can handle set and forget maintenance like feature flag cleanups or post-launch reports.

Anthropic Launches Dynamic Workflows to Orchestrate Fleets of Claude Code Subagents

AnthropicMay 28

Anthropic Launches Dynamic Workflows to Orchestrate Fleets of Claude Code Subagents

Anthropic introduced dynamic workflows in research preview for Claude Code, allowing the tool to write its own orchestration scripts and spawn fleets of parallel subagents. This shift moves planning logic from the model's context window into a dedicated runtime to handle massive migrations and codebase-wide audits.