What happened
Anthropic introduced "Routines" for Claude Code, enabling automated execution of saved configurations comprising a prompt, repositories, and connectors. These routines run autonomously on Anthropic's cloud infrastructure, triggered by schedules, API calls, or GitHub events like pull requests. Routines execute full Claude Code cloud sessions, performing shell commands and utilising skills. Available on Pro, Max, Team, and Enterprise plans, they have daily run limits and operate under individual user accounts, with actions attributed to the user's connected GitHub identity.
Why it matters
Development teams gain automated, unattended execution for repetitive coding tasks, reducing manual effort as routines run on Anthropic's cloud. Procurement teams must account for individual user licensing and daily run caps, as routines tie to personal accounts. Security architects should note routine actions, including commits, appear under the connected user's GitHub identity, requiring careful scope management for repository access and connector permissions. This follows Anthropic's Claude Code Security tool release in February.
Subscribe for Weekly Updates
Stay ahead with our weekly AI and tech briefings, delivered every Tuesday.
