Claude Code adds a /goal command to work until a task is done
By Steven Van ·
Anthropic frames it as part of a broader push to keep Claude Code running to completion rather than stopping after one pass.
The Claude Code CLI now has a /goal command: state what you want done and Claude keeps working until that goal is met, rather than stopping after one pass. Anthropic describes it as one of several recent additions aimed at keeping Claude Code working through to completion on a task. Details on X.
Claude
AI assistant by Anthropic — write, analyze, code, and create with one of the most capable large language models available.
View Claude →