After each ask you type, says what it looked like it needed and what it cost; after a simple or routine ask on a bigger model or effort than it needed, what Haiku, Sonnet or a lower effort would have cost; warns when an ask runs on too small a model or a complex one at low effort; says when the context has grown expensive and how to shrink it; ships /model-fit:baton, a handoff note for a fresh session. Optional Jev shadow labels for comparison.
1 skill
Asks one to three questions about what this session changed, one at a time, and checks your answers against the code.
1 skill
Notices snags that come back across sessions (search spirals, fix loops, large tool outputs) and offers one change to the environment that stops them.
Flags a turn that wrote a lot and offers a /guide reading order; before Claude commits, measures the change's blast radius and asks Jev about risk and readability, then lets it through, flags it, or holds it for review.