🔍
AI EngineeringPython

Idea Critique

by Light0305

Idea Critique is an AI Engineering skill for Claude Code, published by Light0305 in Light-skills.

544 stars69 forkson Light0305/Light-skillsAdded 2026/08/26+2% in starsRepository updated 2026/07/06
academic-writingagent-skillsai-agentai-skillsclaude-codecodexlatexliterature-reviewllm-skillsopencodepaper-writingresearch-toolsscientific-figuresscientific-workflow
Install in seconds
Install Idea Critique
Copy Idea Critique into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/Light0305/Light-skills/tree/master/skills/light-idea-critique ~/.claude/skills/light-idea-critique

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/Light0305/Light-skills.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
skills/light-idea-critique/SKILL.md in Light0305/Light-skills
Installs to
~/.claude/skills/light-idea-critique
Collection
One of 23 skills cataloged from this repository
Category
AI Engineering2451 skills

What Idea Critique does

Idea Critique checks a research idea against prior work, novelty, and fatal flaws using top-conference review standards. Use it when you need a hard yes/no on whether an idea is novel enough and publishable, or when you want the strongest objections before moving on.

Idea Critique is cataloged under AI Engineering on DirSkills. Idea Critique comes from a repository tagged academic-writing, agent-skills, ai-agent, ai-skills and claude-code.

Documentation

README

审 idea(idea-critique)—— 科研主线 stage 4 · 顶会审稿人标准严审 → 一票否决 ⇄ stage 3 重生成

你是 Light 科研流水线的 DAG 第 4 节点。任务不是"打个分鼓励一下",是以顶会审稿人标准严审,把 撞车 / 无创新 / 数据不支撑这类 fatal flaw 一票否决——一个致命缺陷即拒,不被其他高维度平均救回。被毙的 idea 带根因 + 具体缺口 + 最像的前作回 idea-generation(stage 3)定向重生成,构成 3⇄4 双向回环

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Idea Critique

  • What else does Light0305 publish alongside Idea Critique?

    Idea Critique is one of 23 skills that DirSkills catalogs from Light0305/Light-skills, the repository it ships in. Its siblings there include Citation, Data Engineering and Frontend Design. Each one is a separate skill with its own page in this directory, installs the same way Idea Critique does, and is maintained by Light0305 in that same repository. The rest of the collection is listed on the Light0305/Light-skills page.

  • How does Idea Critique compare to other AI Engineering skills?

    Idea Critique ranks #2057 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Idea Critique against them. Open each page to compare what they document and how they install.

More from Light0305/Light-skills

Idea Critique is one of 23 skills cataloged on DirSkills from Light0305/Light-skills.

See all 23 skills
📚
5d ago

Citation

Citation verifies scholarly references and whether a citation actually supports a claim. It is used to audit manuscripts, detect fabricated or chimeric references, and produce a canonical citation registry with provenance-preserving evidence.
Writing
54469
🗂️
5d ago

Data Engineering

Data Engineering checks whether a dataset can support a research idea, including size, quality, license, versioning, and split strategy. It also audits for train-test leakage before analysis or model training.
Data
54469
🎨
5d ago

Frontend Design

Frontend Design turns vague UI requests into runnable React, Tailwind, and shadcn code with design decisions explained. Use it for new interfaces, redesigns, dashboards, landing pages, and accessibility checks.
Frontend
54469
🧭
5d ago

Light Consistency

Light Consistency runs cross-material checks for terminology, metric values, innovation statements, and method names across papers, slides, code, and project docs. Use it before submission or after controlled definition changes to find conflicts and missing coverage without rewriting content.
Quality
54469
🧪
5d ago

Light Experiment Coding

Light Experiment Coding implements frozen research plans as minimal, test-first experiment code with controls for leakage and reproducibility. Use it to build or review training, preprocessing, and evaluation runs with raw artifacts for result analysis.
Quality
54469
📊
5d ago

Light Figure

Light Figure plans and renders publication-ready scientific figures that support specific claims. It is used when turning analysis results into paper figures, checking visual honesty, and reviewing rendered charts for layout issues.
Data
54469