Research Ideation logo

Research Ideation

Community
HughYau
research-ideation

Decide what research problem to attack and how to attack it — the two halves of想idea. Applies a hard four-case novelty gate (cases 1-2 mean a well-established solution exists and the work must not proceed), ranks candidate problems by importance, then derives a pipeline-level method by stripping the surface symptom to the real technical challenge and combining techniques from a challenge-insight tree. Returns a filled scorecard with a go / change-topic verdict, not an opinion, and ships a fully worked example scorecard showing the granularity each cell needs. Use whenever the user floats an idea, picks a direction, asks whether something is novel or already done, asks how to solve a failure case, proposes stacking two existing methods, or asks you to implement a research idea — especially that last one, since implementing an already-solved problem is the most expensive mistake this methodology names. Pairs with literature-review for the retrieval the gate depends on. 中文触发:选题、想idea、这个题值不值得做、有没有创新空间、novelty够不够、做什么方向、怎么解决这个问题、设计方法、设计pipeline、能不能把A和B结合起来、这个方向做完了吗。

Overview

PublisherHughYau
Repositorypengsida-learning-research-skills
Skill nameresearch-ideation
Stars
52
Forks
4
Bundled files
2
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by HughYau on GitHub. Read the source before you install it.

Installation

Install the Research Ideation AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/HughYau/pengsida-learning-research-skills/tree/main/research-ideation
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/HughYau/pengsida-learning-research-skills.git /tmp/pengsida-learning-research-skills
mkdir -p .claude/skills
cp -r /tmp/pengsida-learning-research-skills/research-ideation .claude/skills/research-ideation
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Research Ideation in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Research Ideation on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Research Ideation is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

选题与解题

想 idea 分两步:选一个好问题,然后解决它。 选题决定论文的 novelty 与价值上限,影响远大于后面的解法。 两步共用一个判据,所以放在一个 skill 里。

闸门:四种情况

任何研究动作之前先跑这一步。 把候选 failure case 对号入座:

情况定义结论
1同样输入输出的任务,已有不错方案,只是某些地方不够好换题。创新空间≈0
2输入/输出略变但已有不错方案;或多个不同 data domain、技术内核相同的任务都已有类似方案换题
3只有一两个不同 data domain、技术内核相同的任务有不错方案可做(适合新手)
4各领域虽有类似技术问题,但都没有好方案可做(适合高手)

判定必须基于检索,不能凭印象——判错的代价是几个月。用 literature-review:

python
exec(open("<skills>/literature-review/kernel.py").read())   # <skills> = 本套 skill 的根目录
hits = search_openalex("<failure case 的技术描述>")     # 同类工作
both = expand_citations("<最相关的 DOI>")               # 前后向各走一步引文图

情况 2 和 3 的分界全在"这个技术内核在别的 data domain 上被解决过没有", 关键词检索答不了,必须走引文图。读候选论文的 pipeline 用 pdf-explore (pdf_outline 定位 method 段 → pdf_pages 只取那几页)。

推论:解决重要 failure case 的方法必然 novel。 若简单技术组合就能解决,该 failure case 就不重要;若它足够新颖, 简单组合解决不了,就会逼出新技术。所以 failure case 要尽量 general。

好问题的判据

  • 研究空间大;技术挑战当前学术界够得着;general,限定词少 (反例:只处理"快速运动";正例:处理"复杂动作")
  • 重要程度 = 影响范围 × 迫切需求。影响范围看有多少方向都需要它; 迫切需求看业界是否在反复提这个痛点
  • 不停在"能看"。业界最常提的痛点若只做到勉强可用就转向, 这个方向上等于没有贡献——发了几篇论文但什么问题都没真正解决

三个陷阱:

  1. 做有明显答案的课题(顺利做完、可能中稿、阅读量≈0)
  2. 知道有哪些课题但不排序,想到能做的就动手 → 掉进陷阱 1
  3. 选了对终极目标重要、但技术就绪度不够的问题

怎么找到新的 failure case

"这个任务好像被做完了"的成因:论文只放最好的结果。换数据往往就不 work。

  • 探索算法上限,尝试更具挑战性的 cases(通常是数据)
  • 新的 task setting 或新的数据上最容易发现新 failure case
  • 在新数据上探索方法可能性、给出新的实验结论,本身就是很大的贡献

⚠️ 不要在某技术的原有 setting/数据/failure case 上改进它——空间往往很小。

Goal-driven vs Idea-driven

Idea-driven(Make X work better)Goal-driven(Work for a goal)
起点看到一篇有趣的 SOTA 论文设定目标 → 规划 roadmap → 选 task
风险除非效果好一大截否则影响力有限;易撞 idea;可能被那篇论文的错误路线带偏;创新空间受限于已有解法规划好 roadmap 很难;易出 task-specific technique

默认 goal-driven。 它还有产出保证:追求重要任务、通过 relax 条件总能做出些 work 的结果; 而赌一个新技术能调 work,在实验科学里风险太大。

新锤子出现时:拿它去解决自己 roadmap 上的 milestone task (Transformer→LoFTR、NeRF→Neural Body、SD→DreamFusion), 而不是在新锤子自己的 setting 上刷点——后者仍是 follow-up。


解题:从 failure case 到 pipeline

发现 failure case
      ↓
抽离表面问题,得到真正的 core technical challenge   ← 最常被跳过
      ↓
从 challenge-insight tree(武器库)中选技术,创新性组合

跳过中间那步,得到的是补丁式 trick:既不 general 也说不出 motivation。

三个要点

  1. 针对技术原因在本质上解决问题(保证有效性)
  2. 在 pipeline 层面设计,不是小修小补(否则是 4 类 novelty,reviewer 判 incremental)
  3. 若简单 A+B 就能解决,果断换题——这同时验证了闸门:能被 A+B 解决说明 failure case 不重要

组合是技术创新的常态

先知道有哪些技术、它们在解决什么问题,再组合。做法: 构建 challenge-insight tree → 选技术做创新性组合 → 把可能的 pipeline 都列出来对比再选。

新技术= 组合用在
NeRFOccupancy network + Differentiable renderingreconstruction from images
EG3DStyleGAN + GRAF + Conv occupancy network3D GAN
DreamFusionSDS loss + NeRFtext-to-3D
MVPNeural volumes + Local radiance fieldshuman reconstruction

⚠️ 组合不能是 input → A → 中间输出 → B → output 这种完全拼接。 正常情况下直接拼两个方法也解决不了问题——否则这问题就没有 technical challenge。

方案自查(reviewer 会攻击的点)

  • 针对本质技术原因,不是表面现象
  • pipeline 层面设计,不是给已有 pipeline 打补丁
  • 简单 A+B 达不到同样效果
  • 每个 module 都能说清 motivation 和 technical advantage
  • general,不需要每个场景调超参
  • 提升一个 metric 且不明显损害其他 metrics(净收益为正)

最后一条是 5.4 类拒稿的判据:新设计带来 benefit 的同时引入更强 limitation,收益为负。


输出模板

评估一个 idea 时产出这张卡,不要给散文式意见。 散文里能含糊过去的地方,表格里必须填出来。

markdown
## Idea 评估:<一句话>

**Project setting**:<输入> → <输出>
**failure case**:<具体到能设计实验复现>

### 闸门
**情况 <N>** —— 证据:<哪些论文、哪个 data domain、解决到什么程度;附 DOI>
情况 1/2 → **换题**,并给出 2-3 个替代 failure case

### 问题质量
| 维度 | 判断 | 依据 |
|---|---|---|
| 影响范围 |  |  |
| 业界迫切度 |  |  |
| failure case 是否 general |  |  |
| 技术就绪度 |  |  |
| 三个陷阱 | 未命中 / 命中第 N 个 |  |

### 解法(闸门通过才填)
- **core technical challenge**(抽离表面问题后):
- **候选 pipeline**(≥3 个,含一个故意保守的)

| # | Pipeline | 组合了哪些技术 | 优势 | 劣势/风险 | 只是 A+B? |
|---|---|---|---|---|---|
- **选定 #<N>**,理由:<为什么优于其他两个>

- **逐模块三元素**(直接进 Method section)

| 模块 | Module design(输入→步骤→输出) | Motivation | Technical advantage |
|---|---|---|---|

- **自查 6 条**:<逐条结论>

### 结论
**<做 / 换题 / 需更多检索>** 需检索什么:<具体查询,别猜>

"≥3 个候选、含一个故意保守的"是刻意的:只列一个方案时你是在为已想好的答案找理由。 保守方案提供下界——若它与创新方案效果相当,说明 challenge 不成立。

填之前先看一遍填满的样子:references/scorecard-example.md (3DGS 动态重建遇到运动模糊,一个初看像情况 1、走完引文图后判为情况 3 的例子)。 它示范的是颗粒度——每一格后面都跟着可核对的依据,没有一格写"比较有创新性"这种话。

衔接

要做的事用什么
检索、引文图、DOI 校验literature-review
读候选论文的 pipelinepdf-explore
两棵树怎么建references/literature-tree.md
验证选定方案research-experiment
结论落盘research-project:docs/topic.md、docs/method.md、PROJECT.md §2/§3

PROJECT.md §2 填了情况 1 或 2 时,project_status.py 报阻塞—— 让红灯在每次接手项目时被重新提醒。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Research Ideation AI skill do?

Decide what research problem to attack and how to attack it — the two halves of想idea. Applies a hard four-case novelty gate (cases 1-2 mean a well-established solution exists and the work must not proceed), ranks candidate problems by importance, then derives a pipeline-level method by stripping the surface symptom to the real technical challenge and combining techniques from a challenge-insight tree. Returns a filled scorecard with a go / change-topic verdict, not an opinion, and ships a fully worked example scorecard showing the granularity each cell needs. Use whenever the user floats an...

Why use Research Ideation on TypingMind?

Because you install it once and use it with any model. Research Ideation is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Research Ideation in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/HughYau/pengsida-learning-research-skills/tree/main/research-ideation. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Research Ideation?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Research Ideation?

As many as you like. As long as a model supports skills, you can use Research Ideation with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Research Ideation AI skill free?

It is published on GitHub by HughYau. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇