9.9K ·GitHub
What is this?
Supervisor-Skills is a collection of structured AI skill/prompt files that distill a professor's decade of research experience from top conferences (SIGMOD, VLDB, ICML, NeurIPS) into tools covering the full paper lifecycle, from idea evaluation to writing, figure design, and rebuttal.
Supervisor-Skills is a collection of structured AI skill/prompt files that distill a professor's decade of research experience from top conferences (SIGMOD, VLDB, ICML, NeurIPS) into tools covering the full paper lifecycle, from idea evaluation to writing, figure design, and rebuttal.Is this a code library or something I run programmatically?No, it is not code to run or import. It is a set of plain-text Skill/prompt files you load into an AI assistant such as Claude, GPT-4, DeepSeek, or Kimi so the model follows the embedded research methodology.
No, it is not code to run or import. It is a set of plain-text Skill/prompt files you load into an AI assistant such as Claude, GPT-4, DeepSeek, or Kimi so the model follows the embedded research methodology.How do I install these skills?You copy the quick-start prompt provided in the README and send it to your AI assistant (Claude Code, Cursor, Codex, etc.), and it completes the installation for you.
You copy the quick-start prompt provided in the README and send it to your AI assistant (Claude Code, Cursor, Codex, etc.), and it completes the installation for you.Which AI assistants can run these skills?Any capable LLM chat assistant works; the README specifically names Claude, GPT-4, DeepSeek, Kimi, Claude Code, Cursor, and Codex as compatible.
Any capable LLM chat assistant works; the README specifically names Claude, GPT-4, DeepSeek, Kimi, Claude Code, Cursor, and Codex as compatible.Can I use this for commercial purposes?No. The project is licensed under CC BY-NC-SA 4.0, so sharing and remixing is free only for non-commercial use, and attribution is required.
No. The project is licensed under CC BY-NC-SA 4.0, so sharing and remixing is free only for non-commercial use, and attribution is required.Who is this aimed at?It's aimed at PhD students and early-career researchers in AI/CS who need hands-on guidance on idea evaluation, paper writing, and peer review, not just high-level advice.
It's aimed at PhD students and early-career researchers in AI/CS who need hands-on guidance on idea evaluation, paper writing, and peer review, not just high-level advice.What topics does the accompanying Handbook cover?Six chapters: how to judge paper quality, the idea lifecycle and 5-dimension framework, paper-writing templates (technical and benchmark papers), research figure design, vibe-research practices, and case studies of top-conference papers like Alpha-SQL and AFlow.
Six chapters: how to judge paper quality, the idea lifecycle and 5-dimension framework, paper-writing templates (technical and benchmark papers), research figure design, vibe-research practices, and case studies of top-conference papers like Alpha-SQL and AFlow.How does the Paper Writer skill avoid fabricated citations?It uses an evidence-gated approach where every factual claim in the draft must trace back to your own material or an independently verified source, so citations are checked rather than invented.
It uses an evidence-gated approach where every factual claim in the draft must trace back to your own material or an independently verified source, so citations are checked rather than invented.What individual skills are included?Twelve skills: Idea Evaluator, Vibe Research Guide, Introduction Drafter, Paper Writer, Paper Polish, Deep Research, Tech/Benchmark Paper Templates, Pre-Submission Reviewer, Figure Design Advisor, Draw.io Reconstruction, and Rebuttal Guidance.
Twelve skills: Idea Evaluator, Vibe Research Guide, Introduction Drafter, Paper Writer, Paper Polish, Deep Research, Tech/Benchmark Paper Templates, Pre-Submission Reviewer, Figure Design Advisor, Draw.io Reconstruction, and Rebuttal Guidance.Why did the author create this project?HKUST(GZ) assistant professor Yuyu Luo noticed PhD students repeatedly stuck on the research 'last-mile problem'-knowing theory but lacking judgment and taste-so he distilled his decade of top-venue publishing and reviewing experience into executable AI skills.
HKUST(GZ) assistant professor Yuyu Luo noticed PhD students repeatedly stuck on the research 'last-mile problem'-knowing theory but lacking judgment and taste-so he distilled his decade of top-venue publishing and reviewing experience into executable AI skills.What are similar projects?Research-Starter-Kit, awesome-ai-research-writing, PaperSpine, academic-research-skills, learning_research, nature-skills, Evidence-Bound-Press-Conference-Revision-Skill, open-science
Research-Starter-Kit, awesome-ai-research-writing, PaperSpine, academic-research-skills, learning_research, nature-skills, Evidence-Bound-Press-Conference-Revision-Skill, open-scienceWhich topics is it in?research-tools, ai-research, academic-writing, deep-research, writing, paper, academic, academic-research, awesome, skills, guide, awesome-list
research-tools, ai-research, academic-writing, deep-research, writing, paper, academic, academic-research, awesome, skills, guide, awesome-list