fixed_juice_ralph

ralph_loop with a governor on it: a fresh session every round, and between the rounds the effort moved a rung to hold the agent to juice output tokens per turn of the model.

hmz exec -f fixed_juice_ralph -a claude/claude-opus-5:high "$(cat TASK.md)"

fixed_juice_ralph

A governor rather than a brake: what it holds steady is the size of an answer. Under the target the agent is asked to think harder, over it to think less — one rung of its own model’s ladder per round.

A governor is not a brake

What it holds steady is the size of an answer. A loop held to 2000 output tokens a turn goes on at 2000 a turn for as long as anybody leaves it running; what ends it is the run's allowance. juice and the allowance are one quantity read at two scales — what a turn is worth, and what the run is.

Per turn of the model — one request and the answer to it — rather than of the flow, which is many of those plus whatever the tools took. That average is what an effort moves: a model asked to think harder writes more in each answer. So an agent under the target is moved up a rung of its own model's ladder and one over it down, once a round, so the loop settles rather than swings.

Nothing here is a clock. How long a round takes and what it costs an hour are what the model and the work make of it.

What it takes

juice: 2000     # output tokens an average turn of the model is to come out with
over: 300       # how far back that average is taken, in seconds
slack: 0.15     # how far off it may be before the effort moves
rest: 5         # seconds between rounds

A slack of 0.15 leaves the effort alone between 85% and 115% of the target. What ends the run is not among these: that is the run's allowance, set under budget: in the same file or on the budget row in /flow, and this flow declares ten million output tokens as what a run of it is worth by default.

What it keeps

rounds and effort — the rung the governor settled at. A loop started again at the top of the ladder would walk back down to that rung one paid turn at a time.

The rung is kept as the effort's own word, not as a place on the ladder: the ladder is whatever the account says its CLI runs today, and a model retired or an effort added moves the places. A word no longer on the ladder reads as no rung, and leaves the agent where a first run would start it.

The answer size is not kept: it is an average over the last few minutes of turns, and a run starting today has none of yesterday's to average.

See also

• Efforts — what a rung is, and which ladder each model has
• Cost and rate — reading what a run is spending while it runs
• ralph_loop — the same loop, ungoverned