feat(appkit): agent eval framework, judge, mlflow connector (stack 2/5) - #478

Open

MarioCadenas wants to merge 8 commits into

mainfrom

pr/agent-evals-2-framework

Commits on Aug 21, 2026

feat(appkit): add agent eval framework with native mlflow evaluation runs
MarioCadenas
committed
4e2435a
feat(appkit): add llm-as-judge and mlflow trace/eval parity
MarioCadenas
committed
da9a4d2
refactor(appkit): add mlflow connector with profile-based oauth
MarioCadenas
committed
0add2e2
docs(appkit): regenerate API reference after rebase onto main
MarioCadenas
committed
bf8be0d
chore(appkit): conform evals + mlflow connector to oxc toolchain
MarioCadenas
committed
d60fad5
refactor(appkit): drop unwired eval config and use recursive readdir
MarioCadenas
committed
f5915d3
refactor(appkit): discover evals from server/agents to match folder-per-agent layout
MarioCadenas
committed
4c6983e
docs(appkit): reconcile API reference after rebase onto main
MarioCadenas
committed
65de478