菜单
Stage ml: ML venue lens (default NeurIPS) scoring per reference for technical fit: canonicality, baseline relevance, recency for SOTA claims; writes ml_* columns.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/jon-chun/cite-guard --skill cite-guard-ml命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | cite-guard-ml |
| description | Stage ml: ML venue lens (default NeurIPS) scoring per reference for technical fit: canonicality, baseline relevance, recency for SOTA claims; writes ml_* columns. |
| license | MIT |
| metadata | {"version":"1.0.0","category":"research-qa","tags":["ml","neurips","icml","iclr"]} |
Per reference:
Default profile: neurips (override via CLI).
Stage audit: bibliography hygiene checks per reference and writes audit_quality/audit_confidence/audit_remediation per row.
Stage ground: evaluates whether each reference supports the claims it is cited for; fetches evidence preferring md/html/tex/rtf/txt/pdf; writes ground_* columns and produces grounding reports.
Initialize the reference QA pipeline by creating out/audit_references.csv with one row per BibTeX entry and required stage columns.
