Claude Code scheduled tasks & /loop
Official Claude Code scheduled prompt and /loop documentation.
Visit sourceSource type: official-docs. License: not stated or not applicable. The linked source supports the visible capability notes but does not prove production suitability.
Open evidence sourceConfirm current maintenance, permissions, data handling, license obligations, failure behavior, and a human approval gate in your own environment.
Practical fit
Scheduled tasks and /loop are useful for small observations that become valuable when repeated: checking a queue, reviewing fresh failures, or producing a regular status note. They are less suitable for open-ended coding work where each run can expand the scope.
A schedule multiplies both mistakes and cost. The first runs should therefore stay manual until the prompt, output, retry behavior, and ownership of the result are predictable.
How to evaluate it
Example decision
Run a weekday check that reads the latest CI failures and writes a short triage note. Do not let that scheduled check edit code until several manual runs show that the discovery and classification steps are reliable.
Verify before you adopt it
Source review note
We reviewed the primary source on July 21, 2026. Product behavior, pricing, licenses, and maintenance status can change, so check the linked source before adopting it.