Prompt evals over MCP: run a prompt on your dataset, score each output 1-5 with an LLM judge.

Описание

Change your AI with confidence, not crossed fingers. · Prompt evaluation for AI apps. Run prompts against real inputs, score outputs 1-5 with an LLM judge, ship only what improved. Free on Cloud, or self-host. · You changed the prompt. It feels better. You need to know. · Four moves. Then you have evidence. · Point your coding agent at CompletionKit. Walk away. · What you actually get, line by line.

Установка

Не знаете, с чего начать — попросите ассистента провести по шагам:

Характеристики

Тип	MCP-сервер
Категория	MCP-серверы
Цена	бесплатно
Платформа	Командная строка
Системы	cli, api
Хостинг	cloud
Установка	mcp
Протоколы	mcp
Язык сайта	en
GitHub	homemade-software-inc/completion-kit
Запуск	2026-07-18