LangChain shipped an eval-engineering skill that interviews you instead of one-shotting evals.

The July 22 post describes an installable skill that reads your repo to map the agent surface (prompts, models, tools, skills, hooks), pulls production traces via langsmith-cli, then interviews you to converge on evals you approve one at a time (LangChain). Output is a containerized Harbor task per eval under evals/<task-id>/ with instruction.md, an environment/ Dockerfile, and tests/ verifier logic. The reported lesson: single-pass generation produced weak evals, and the good ones came from inspecting both the agent's and the verifier's trajectories to catch reward-hacked shortcuts.