Adds Dokimos to the Artificial Intelligence section.

The other entries in this category build LLM or agent apps. LangChain4j, JamJet, and the MCP Java SDK all help you create them. Dokimos does the opposite: it tests them. It scores responses, validates tool calls and execution traces, and fails the build on a quality regression, all as a normal JUnit test. It is the only evaluation framework in the category.

Disclosure: I maintain Dokimos. It is MIT licensed and published to Maven Central under dev.dokimos. I think it fills a real gap in this list, since none of the AI entries cover evaluation or regression testing.

Summary by cubic

Add Dokimos to the Artificial Intelligence section to cover evaluation and regression testing for LLM and agent apps. It scores responses, validates tool calls and execution traces, and catches quality regressions in CI.

^{Written for commit 7eeba08. Summary will update on new commits.}