budget
CommandMarks the cheapest Claude model and effort that still clears this task with high confidence. Still shows the full Budget,…
Effort recommendation only for this task. Shows a narrow ladder of just the effort level, from a lower setting to a higher one, for a model named in the task or else the one you are on.
$ npx -y skills add JordanGrothentic/caliber --agent claude-code
description: Effort recommendation only for this task. Shows a narrow ladder of just the effort level, from a lower setting to a higher one, for a model named in the task or else the one you are on. argument-hint: [describe the task] disable-model-invocation: true
Locate the caliber skill files, trying these locations in order: the project `.claude/skills/caliber/`, then `~/.claude/skills/caliber/`, then a Glob for `**/skills/caliber/SKILL.md`. Read `SKILL.md` and `models.md` from the first location that has them. If they cannot be found, stop and tell the user the caliber skill is not installed (copy `skills/caliber` from the repo into `~/.claude/skills/` or the project `.claude/skills/`), and never recommend models, efforts, or prices from memory.
Apply the engine to the task below and produce the effort only ladder exactly as the skill's Entry points section defines for `/which-effort`; the engine is authoritative. Choose the model to ladder on in this order, and always name which one you used: a model named in the task text, else the session's current model (you know which model you are running as), else the Optimal pick.
Task: $ARGUMENTS
Right size every Claude task. Caliber is a set of Claude Code slash commands, plus an uploadable Claude skill for the website, that tells you which model and effort level to use for the task in front of you, instead of defaulting to the biggest model at the
Marks the cheapest Claude model and effort that still clears this task with high confidence. Still shows the full Budget,…
Marks the strongest Claude model and effort that genuinely helps this task, for high stakes or hard work. Still shows the full…
Model recommendation only for this task. Shows a narrow ladder of just the model, from a cheaper option to a stronger one, at one…
Implement tasks incrementally — build, test, verify, commit. Add "auto" to run the whole plan in one approved pass.