Design evaluation datasets and supporting documentation for AI systems and agents, covering the scoping interview, difficulty distribution, dataset contract, sample review, and metric and tooling selection. Use when building or reviewing an evaluation set for a conversational agent, assistant, or retrieval-grounded AI system.
Install
npx skills add microsoft/hve-core --skill evaluation-designnpx skills add microsoft/hve-core --skill evaluation-design -a claude-code -gnpx skills add microsoft/hve-core --skill evaluation-design -a codex -ggemini skills install https://github.com/microsoft/hve-core.git --path .github/skills/data-science-engineering/evaluation-designgh skill install microsoft/hve-core evaluation-designgit clone --depth 1 https://github.com/microsoft/hve-core /tmp/microsoft__hve-core
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/microsoft__hve-core/.github/skills/data-science-engineering/evaluation-design ~/.claude/skills/evaluation-design # Claude Code
cp -R /tmp/microsoft__hve-core/.github/skills/data-science-engineering/evaluation-design ~/.agents/skills/evaluation-design # Codex, Cursor, Gemini CLI, Copilot, OpenCode