research-gap
Skill589 starsChanged yesterdayMITView on GitHub
Use when the user wants to identify research gaps, open questions, unexplored areas, cross-paper limitations, or future directions from workspace literature.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- devinat1/skillstheir own version, first seen Oct 6, 2026
- DJDeborah/research-mechanics-skillstheir own version, first seen Oct 2, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- devinat1/skills★ 1same name
- DJDeborah/research-mechanics-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository47 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/scholaraio.mdcCursor rule
- academic-writingSkill
- arxivSkill
- auditSkill
- backupSkill
- bioinformaticsSkill
- citation-checkSkill
- citationsSkill
- documentSkill
- drawSkill
- enrichSkill
- exploreSkill
- exportSkill
- graphSkill
- gromacsSkill
- importSkill
- indexSkill
- ingestSkill
- insightsSkill
- lammpsSkill
- literature-reviewSkill
- metricsSkill
- nature-workflowSkill
- openfoamSkill
- paper-guided-readingSkill
- paper-writingSkill
- patent-fetchSkill
- patent-searchSkill
- posterSkill
- publishSkill
- quantum-espressoSkill
- renameSkill
- review-responseSkill
- scientific-runtimeSkill
- scientific-tool-onboardingSkill
- scrubSkill
- searchSkill
- setupSkill
- showSkill
- technical-reportSkill
- topicsSkill
- translateSkill
- workspaceSkill
- writing-polishSkill
The file
Files
研究空白识别
从工作区文献中系统性地发现研究空白和开放问题。
前提
用户必须指定一个 workspace(--ws NAME),且工作区中应有足够数量的论文(建议 10+ 篇)。
报告语言由用户指定(中文 / English)。
执行逻辑
1. 全局扫描
scholaraio ws show <name> # 论文列表
scholaraio topics # 主题聚类(如已建模)对工作区论文做 L2 扫描(标题 + 摘要),建立领域地图。show 命令会自动展示已有的 notes.md 历史笔记,优先复用已有发现。
2. 多维度分析
维度 1:主题覆盖度
scholaraio topics # 全库主题分布
scholaraio topics --topic <ID> # 各主题下的论文工作区论文是否集中在某几个主题?哪些相关主题缺乏覆盖?
维度 2:时间趋势
按年份统计工作区论文分布,识别:
- 哪些方向在近几年论文激增(热点)
- 哪些方向论文渐少(可能已成熟或被放弃)
- 哪些方向有早期工作但近年无跟进(潜在空白)
直接用 Python 读取 configured papers library 下的 */meta.json 做统计分析。
维度 3:方法论对比
扫描工作区论文的方法部分(L3-L4),绘制方法论矩阵:
scholaraio show <paper-id> --layer 3 # 结论中通常提及方法- 哪些方法被广泛使用?
- 哪些方法组合尚未被尝试?
- 某方法在 A 问题上成功,是否可迁移到 B 问题?
维度 4:引用图谱空洞
scholaraio shared-refs "<id1>" "<id2>" # 共同引用
scholaraio refs "<id>" # 参考文献
scholaraio citing "<id>" # 被引论文- 哪些论文互相引用但观点矛盾?(未解决的争议)
- 哪些论文被大量引用但缺乏后续验证/复现?
- 哪些关键参考文献不在工作区中?(可能是盲区)
维度 5:论文自述的 Future Work
加载工作区中高引论文的 L3(结论),提取作者自己提出的未来方向:
scholaraio show <paper-id> --layer 3这些 future work 是否已有人做了?交叉搜索验证:
scholaraio usearch "<future work 关键词>"3. 输出报告
生成结构化的研究空白报告,保存到 workspace/<name>/research-gaps.md。
对每个发现的空白,按类型分类:
| 空白类型 | 含义 | 示例 |
| 知识空白 | 某个现象/问题尚无人研究 | "高 Re 下的 X 效应尚未被测量" |
| 方法空白 | 已有结论但方法存在缺陷或局限 | "现有研究均为 RANS,缺乏 DNS 验证" |
| 矛盾空白 | 不同研究给出矛盾结论 | "A 组报告正效应,B 组报告负效应" |
| 迁移空白 | 某方法/发现尚未被推广到相关领域 | "该方法在 2D 有效,3D 尚未尝试" |
| 规模空白 | 只有小规模/受限条件的结果 | "仅低 Re 数据,工程 Re 下未验证" |
报告结构:
- 领域现状概述(2-3 段)
- 已识别的研究空白(按优先级排序)
- 空白类型 + 描述
- 支撑证据(哪些论文暗示了这个空白)
- 潜在研究问题
- 可行性评估(数据/方法/资源是否可及)
- 未解决的争议(如有)
- 建议的下一步
量化辅助:必要时编写 Python 代码从 meta.json 批量提取数据,做统计图表(年份分布、方法频次、参数范围覆盖等),用可视化支撑空白发现。
分析笔记持久化:对深度分析过的论文,必须通过 CLI 将关键发现写入笔记:
scholaraio show "<paper-id>" --append-notes "## YYYY-MM-DD | <workspace> | research-gap
- 关键发现"4. 互动讨论
报告生成后,主动与用户讨论:
- 哪些空白与用户的研究方向最相关?
- 是否需要补充更多文献来验证某个空白?
- 能否将某个空白细化为具体的研究问题和假设?
示例
用户说:"帮我看看 drag-review 工作区里还有什么研究空白" → 全面扫描文献,多维度分析,生成研究空白报告
用户说:"这些论文的 future work 都提了什么方向?有人做了吗?" → 提取各论文结论中的 future work,交叉搜索验证
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

