home/categories/lab-tools/hamelsmu-evals-skills-skills-validate-evaluator-skill-md
lab-toolsresearch

validate-evaluator

Calibrate an LLM judge against human labels using data splits, TPR/TNR, and bias correction. Use after writing a judge prompt (write-judge-prompt) when you need to verify alignment before trusting its outputs. Do NOT use for code-based evaluators (those are deterministic; test with standard unit tests).

hamelsmu
maintainer
hamelsmu
Обновлено 3/3/2026
Звёзды
1105
Форки
117
quick start

Installation and usage

Calibrate an LLM judge against human labels using data splits, TPR/TNR, and bias correction. Use after writing a judge prompt (write-judge-prompt) when you need to verify alignment before trusting its outputs. Do NOT use for code-based evaluators (those are deterministic; test with standard unit tests).

Установка
$ install --globalskills.sh
Использование

После установки вы можете использовать этот skill, выполнив следующую команду в терминале:

skills use validate-evaluator