Skip to main content

​
clawker loop

Run Claude Code in autonomous loops

​
Synopsis

WARNING: This command is experimental and may change in future releases. Commands for running Claude Code agents in autonomous loops. The loop command automates Claude Code execution: Claude runs repeatedly until signaling completion via a LOOP_STATUS block in its output. Two loop strategies are available: iterate Same prompt repeated fresh each invocation tasks Agent reads a task file, picks an open task, does it, marks it done Container lifecycle is managed automatically — a fresh container is created for each iteration and destroyed afterward. Workspace and config volumes persist across iterations so the agent sees cumulative codebase changes. Available commands: iterate Run an agent loop with a repeated prompt tasks Run an agent loop driven by a task file status Show current session status reset Reset the circuit breaker

​
Examples

  # Run a loop with a repeated prompt
  clawker loop iterate --prompt "Fix all failing tests"

  # Run a task-driven loop
  clawker loop tasks --tasks todo.md

  # Check the status of a loop session
  clawker loop status --agent dev

  # Reset the circuit breaker after stagnation
  clawker loop reset --agent dev

​
Subcommands

​
Options

  -h, --help   help for loop

​
Options inherited from parent commands

  -D, --debug   Enable debug logging

​
See also

  • clawker - Manage Claude Code in secure Docker containers with clawker