dokimos-dev/dokimos

Create evaluation datasets in JSON, CSV, or JSONL format for Dokimos

Scaffold a new Evaluator implementation following Dokimos conventions

Scaffold a Dokimos Experiment with dataset, task, evaluators, and reporter

Scaffold eval-driven tests using dokimos-junit with @DatasetSource and Assertions.assertEval()

Set up evaluation of AI agents with tool call validation, correctness checks, and task completion using Dokimos

Set up evaluation of Embabel agents using Dokimos

Set up evaluation of Koog AI agents using Dokimos

Set up evaluation of LangChain4j applications and RAG pipelines using Dokimos

Set up evaluation of an OpenAI Java SDK tool-calling agent using Dokimos

Set up evaluation of Spring AI applications using Dokimos

Set up evaluation of Spring AI Alibaba graph agents using Dokimos