AI Agent Hub
Back to skills
NSFC Full Proposal Writing Advisor icon

NSFC Full Proposal Writing Advisor

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Follow https://skillhub.cn/install/skillhub.md to install @user_5933f5b3/nsfc-full-package into your AI assistant.

About this skill

What It Solves

NSFC proposals often fail not because of length, but because reviewers cannot trace the logic: the title is broad, the scientific question stays at the phenomenon level, the hypothesis reads like correlation, the research content mixes tools and databases, prior work looks like a CV, and the abstract is filled with generic importance claims. NSFC-full-package turns these review concerns into concrete writing constraints, pushing the proposal from “what is happening” to “what mechanism drives it.”

How It Works

  • Question sharpening: blocks descriptive verbs such as explore, describe, and summarize; favors mechanism terms such as evolve, drive, interact, and constrain; outputs three versions centered on mechanistic evolution, interaction network, and boundary response.
  • Hypothesis construction: converts the question into an “upstream trigger → core mechanism → downstream phenotype” causal chain, requiring causal direction, specific nodes, boundary conditions, and falsifiability.
  • Content and plan split: research content keeps scientific variables, mechanisms, and subquestions; research plan carries procedures, tools, and parameters, preventing content from sounding like methods.
  • Prior-work defense: papers, preliminary experiments, and technical accumulation must map to hypothesis nodes and explain how they support the proposal, instead of listing impact factors or grants.
  • Abstract polishing: organizes about 400 words into five sentences—background, gap, hypothesis, content, and significance—while the content sentence keeps scientific variables and logical progression only.

Boundaries

It is useful when you already have a research idea and need to reshape a draft into a complete, review-ready structure. It does not replace experimental design, data verification, or domain judgment. If the question lacks a mechanistic anchor, the skill will tend to ask for conditions, nodes, and causal direction rather than producing a generic title.

Use Cases

  • Sharpen a broad title into reviewable scientific questions
  • Expand hypothesis into upstream, mechanism, phenotype chain
  • Separate scientific variables from tools and procedure steps
  • Map prior papers and pilots to hypothesis nodes with support

Best For

  • Early-career researchers drafting an NSFC proposal structure
  • Advisors reviewing team proposals for hypothesis-content mix
  • Grant writers polishing abstracts, prior work, hypotheses
  • Cross-field researchers linking prior work to mechanism