- Home
- Skills
- Huangwb8
- Chineseresearchlatex
- Nsfc Length Aligner
nsfc-length-aligner_skill
- Python
1.3k
GitHub Stars
4
Bundled Files
2 months ago
Catalog Refreshed
3 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-length-aligner- CHANGELOG.md4.5 KB
- config.yaml4.8 KB
- README.md2.6 KB
- SKILL.md7.6 KB
Overview
This skill quantifies and aligns the length of NSFC proposal manuscripts to budgeted page/character targets and provides focused rewrite suggestions. It detects which files or sections are over- or under-budget and produces action lists plus meaning-preserving expansion or compression to meet requirements. The tool emphasizes measurable, repeatable checks and a closed-loop edit → recheck workflow.
How this skill works
The tool parses the LaTeX project tree (following main.tex includes) or a compiled PDF and measures per-file and per-section character and page counts against a configured length standard. It outputs a human-readable report and machine-readable JSON showing deltas, section contributions, and ranking of the largest deviations. Based on the report, it recommends targeted edits and can perform conservative, meaning-preserving expansions or compressions to hit the budget without changing the argument flow.
When to use it
- Before finalizing an NSFC submission to verify page and character budgets.
- When certain proposal parts feel subjectively too short or too long and need quantification.
- To create a prioritized edit plan targeting the largest deviations first.
- After edits, to re-run and confirm the proposal now meets page/character constraints.
- When you need to expand or compress text while preserving the original meaning and evidence chain.
Best practices
- Set the length standard to the current year’s guideline in the configuration before running checks.
- Prefer passing the compiled PDF for accurate page counts and use character counts for deterministic rewrite loops.
- Read the generated length_report.md and JSON before editing to localize changes to 1–2 offending sections when possible.
- When expanding, add verifiable definitions, hypotheses, experimental contrasts, and risk mitigation rather than generic filler.
- When compressing, remove repetition and shrink background; preserve the problem-method-verification sequence.
- Always re-run the check after edits to complete the edit → recheck closed loop.
Example use cases
- You suspect the Research Content section is too short—run the checker to quantify and get section-level targets.
- Your compiled proposal hits 31 pages—identify which files contribute most and compress targeted subsections to reach ≤28 pages.
- You need to pad a short 'Research Basis' section with evidence-linked details without adding new claims.
- Create an ordered action list to fix the top 3 files with the largest |delta| before deadline.
- Validate that post-edit changes kept argument flow intact by re-running the length audit.
FAQ
It uses a configurable length_standard; an example aligned to 2026 guidance is provided (30% basis / 50% content / 20% foundation and suggested ≤28 pages). Update the config to match the active year’s template rules.
Can it edit my proposal automatically?
Yes—automated edits aim to preserve meaning and minimize change. Edits are conservative: expansions add verifiable detail and compressions remove repetition or background. Always review edits and re-run the checker.