nsfc-grant-writing

name	nsfc-grant-writing
description	Use when the user writes, revises, reviews, or plans a Chinese NSFC grant application, National Natural Science Foundation proposal, guo-zi-ran, 国自然, 基金本子, 立项依据, 科学问题, 研究内容, 技术路线, 创新点, 可行性分析, or grant review response.

NSFC Grant Writing

Purpose

Use this skill for National Natural Science Foundation of China proposal planning, drafting, review, and revision. The bundled folders are preserved from the original kit and can hold future examples, references, outputs, and proposal drafts.

When To Use

• The user mentions NSFC, National Natural Science Foundation, guo-zi-ran, 国自然, 青基, 面上, 重点项目, 基金申请书, or 基金本子.
• The user asks for 立项依据, 科学问题, 研究内容, 研究目标, 技术路线, 创新点, 可行性分析, 年度计划, or 摘要.
• The user wants reviewer-style critique of a grant draft.

Workflow

1. Ask for project type, discipline code, applicant stage, title or rough direction, and any existing draft.
2. Separate the application into four layers:
   • scientific problem
   • hypothesis or core mechanism
   • research aims
   • methods and feasibility evidence
3. Before drafting, identify the weakest link: novelty, logic chain, preliminary data, feasibility, or alignment with funding scope.
4. Write or revise one section at a time. Keep claims specific and evidence-backed.
5. Review as a skeptical panel member: significance, originality, feasibility, team basis, and risk control.

Section Checklist

Section	What To Check
立项依据	Does it move from field need to precise scientific problem?
科学问题	Is it answerable, non-trivial, and not just an engineering task?
研究内容	Are aims independent but logically connected?
技术路线	Can a reviewer see inputs, methods, outputs, and fallback plans?
创新点	Is novelty mechanistic or conceptual, not only "first to apply"?
可行性	Are preliminary data, platform, team, and methods credible?

Local Resources

• proposals/: store proposal drafts or examples.
• references/: store policy notes, writing guides, review rubrics, or discipline-specific material.
• output/: store generated or revised proposal sections.
• dist/: store packaged versions if needed.

Guardrails

• Do not invent policy details, deadlines, funding rules, or discipline-code requirements. Verify current rules when the user asks for official requirements.
• Do not fabricate preliminary results or citations.
• Make uncertainty explicit when the proposal lacks data or field context.