nsfc-justification-writer_skill

This skill helps you craft NSFC project justification by outlining value, current gaps, core questions, and entry point with a theory-driven approach.
  • Python

1.3k

GitHub Stars

4

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill huangwb8/chineseresearchlatex --skill nsfc-justification-writer

  • CHANGELOG.md1.1 KB
  • config.yaml7.2 KB
  • README.md8.2 KB
  • SKILL.md12.7 KB

Overview

This skill automates writing and refactoring the NSFC-style "立项依据" section for research grant proposals. It focuses on making a clear argument for why the work must be done now and preparing a tight transition into the research content. The default orientation is theoretical innovation, with configurable styles for mixed or engineering emphases.

How this skill works

The tool reads the project workspace and updates only the designated LaTeX target file while preserving section headings and template integrity. It guides writing in incremental stages (skeleton → draft → revise → polish → accept), enforces dimension coverage (value, gaps, hypothesis, cut-in), flags risky boastful phrasing, and checks cross-section term consistency. AI-enabled features offer semantic checks and hypothesis framing; if AI is unavailable, robust heuristics and fallbacks ensure useful output.

When to use it

  • When you explicitly ask to write or rewrite an NSFC-style "立项依据" (project justification).
  • When you need a theoretically focused justification that emphasizes falsifiable hypotheses.
  • When preparing or polishing a grant application that must align with a fixed LaTeX template.
  • When you want progressive, auditable edits to only one subsection at a time.
  • When you require checks for risky or unsubstantiated claims before submission.

Best practices

  • Provide the minimal information form (project title, objectives, key references, expected contributions) before generation.
  • Keep section headings intact; request structural changes explicitly and the tool will first propose a unified 4-paragraph skeleton.
  • Supply DOIs or bib entries if you want in-text \cite{} inserted; the tool will not invent verifiable citations.
  • Use the staged workflow: generate skeleton, fill missing info, iterate on one subsection per round, then finalize.
  • Enable terminology configuration so cross-section consistency checks can surface precise metric and acronym mismatches.

Example use cases

  • Transform bullet notes on problem importance into a concise "value and necessity" paragraph suited to NSFC reviews.
  • Refactor an existing justification to replace vague claims with a falsifiable core hypothesis and clear verification dimensions.
  • Integrate a systematic literature-review export as read-only input to build the "domestic and international status" paragraph.
  • Automatically detect and suggest rewrites for absolute or boastful expressions that may harm review credibility.
  • Produce a backup and diff for each applied subsection change to support collaborative revision and rollback.

FAQ

No. It writes only the designated justification tex file and never alters main template/class/style files.

Can it insert citations automatically?

It will add \cite{} only when you provide valid bib entries or DOIs; otherwise it prompts for bibliographic data.

What if the AI responder is not available in the environment?

The tool falls back to deterministic heuristics and hard-coded guidance so work can continue without external model access.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational