loop-optimizer — bounded retroactive loops

Package scope and prerequisites

This catalog ships instructions, not the executable runtime. Python modules, scripts, local services and optional tools named below must be installed separately from the upstream project. Check their availability before running commands; if they are missing, explain the prerequisite instead of claiming execution or results. The host agent may use model tokens while following these instructions.

Purpose

Decides, from observable features, what the loop should do next and when to stop, and can replay a recorded baseline to quantify the difference.

python -m skills.loop_optimizer.cli <action> ...
python -m skills.loop_optimizer.cli replay
python -m skills.loop_optimizer.cli stats

When NOT to use

• You need the context a loop carries between steps — use context_windows.
• You want the loop's knowledge distilled — use agent_intel.
• You want a single task routed — use nn_router or local_router.