Individual skill
light-idea
Generate, challenge, and prioritize evidence-grounded research paper ideas, especially after light-literature. Use for 论文 idea、选题、从精读笔记找研究问题、创新性查重、已有想法打磨、研究方向取舍 and minimal pilot planning. Connect prior work to a precise contribution and a discriminating test. Pure literature retrieval, manuscript polishing, and running full experiments belong to other workflows.
light-idea/SKILL.md
Source commit 6597430ad09d8f7c3518ec4cb08829be2b417e18
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Network or broad shell
- Last checked
- Nov 10, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 908
- Repository last updated
- Nov 8, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- academic-writing Repository topic
- ai-agent Repository topic
- claude-code Repository topic
- codex Repository topic
- llm-skills Repository topic
- paper-writing Repository topic
- research-tools Repository topic
- scientific-workflow Repository topic
- agent-skills Repository topic
- ai-skills Repository topic
- research Derived category
- testing Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
.gitignoreasset · 63 bytesSKILL.mdcore · 12,281 bytesSKILL.mdcore · 12,281 bytesSKILL.mdcore · 12,281 bytesSKILL.mdcore · 12,281 bytesagents/openai.yamlasset · 361 bytesevaluations/2026-11-06-upgrade.mddoc · 6,848 bytesevaluations/scenarios.mddoc · 2,960 bytesevaluations/validation.mddoc · 3,884 bytesreferences/artifact-contract.mddoc · 7,691 bytesreferences/ideation.mddoc · 5,929 bytesreferences/novelty.mddoc · 5,132 bytesreferences/pipeline.mddoc · 6,117 bytesreferences/project-layout.mddoc · 5,991 bytesreferences/repository-audit.jsonasset · 3,332 bytesreferences/revisions-and-handoffs.mddoc · 7,035 bytesreferences/selection-and-tests.mddoc · 6,000 bytesreferences/sources.mddoc · 16,650 bytesscripts/idea_bridge.pyruns · 39,880 bytestemplates/decision.mddoc · 1,041 bytestemplates/idea-card.mddoc · 1,597 bytestemplates/ideas.example.jsonasset · 1,526 bytestests/test_bridge.pyruns · 18,688 bytestests/test_revision_regressions.pyruns · 11,564 bytes
Read excerpt
# Light Idea
## 项目目录与续作
需要落盘或继续已有项目时，先读 [项目目录约定](references/project-layout.md)，按其中的本阶段位置、版本与关联规则保存；用户指定路径和旧项目优先。包内资源相对本 skill，研究产物相对真实项目，二者不要混用。只做局部讨论时无需初始化目录。
把已经读到的文献变成**值得回答的问题、相对已有工作的具体差异，以及最先该做的验证**。默认用中文交流，保留原始论文标题和检索术语。用户只要讨论时直接在聊天中完成；需要可追溯的项目产物时使用下面的工作目录。
宿主模型负责阅读、推理、检索判断和取舍；附带脚本只导入材料、保存指纹和检查关系，不调用模型、不联网、不判定科学创新性。
续作时先恢复当前问题、已采用候选和版本：有 `Memory/` 就读 current 及相关条目来源，按需调用记忆的 context；有 `Workflow/` 就核对本轮任务、已做工作和未决项，再读真实 Idea 记录。特别查看已否决理由、待确认前提和重启条件，避免换会话又推荐同一条已被推翻的路。记忆中的判断遇到冲突或新证据时回来源核对；明确的局部讨论无需初始化 Memory / Workflow。
## 入口与工作量
