Dedupe and rank a raw paper set (`papers/papers_raw.jsonl`) to produce `papers/papers_dedup.jsonl` and `papers/core_set.csv`. **Trigger**: dedupe, rank, core set, 去重, 排序, 精选论文, 核心集合. **Use when**: 检索后需要把广覆盖集合收敛成可管理的 core set（用于 taxonomy/outline/mapping）。 **Skip if**: 已经有人手工整理了稳定的 `papers/core_set.csv`（无需再次 churn）。 **Network**: none. **Guardrail**: 偏 deterministic；输出应可重复（稳定 paper_id、字段规范）。
Install
npx skills add WILLOSCAR/research-units-pipeline-skills --skill dedupe-ranknpx skills add WILLOSCAR/research-units-pipeline-skills --skill dedupe-rank -a claude-code -gnpx skills add WILLOSCAR/research-units-pipeline-skills --skill dedupe-rank -a codex -ggemini skills install https://github.com/WILLOSCAR/research-units-pipeline-skills.git --path .codex/skills/dedupe-rankgh skill install WILLOSCAR/research-units-pipeline-skills dedupe-rankgit clone --depth 1 https://github.com/WILLOSCAR/research-units-pipeline-skills /tmp/WILLOSCAR__research-units-pipeline-skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/WILLOSCAR__research-units-pipeline-skills/.codex/skills/dedupe-rank ~/.claude/skills/dedupe-rank # Claude Code
cp -R /tmp/WILLOSCAR__research-units-pipeline-skills/.codex/skills/dedupe-rank ~/.agents/skills/dedupe-rank # Codex, Cursor, Gemini CLI, Copilot, OpenCode