taskcut

Compacts when the work moves on from one sub-task to the next instead of when the context window fills, including in the middle of a long turn. Once the context is past a floor, a model judges each step, reading what you asked for and where Claude has got to but never any tool output, for whether the work moves on from a finished piece to another; if it does, taskcut runs Claude Code's own compaction. Below the floor it does nothing and costs nothing.

Benchmark harness, not a plugin to install: runs taskcut's judge over labelled steps.

Benchmark harness, not a plugin to install: records the judge's prompt at every step of a live session, to check the replay against.