nsfc-justification-writer

当用户明确要求"写/改 NSFC 立项依据""立项依据写作/重构"时使用。基于最小信息表输出价值与必要性、现状不足、科学问题/假说与项目切入点，并保持模板结构不被破坏。适用于 NSFC 及各类科研基金申请书的立项依据写作场景。

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

85%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill body is a well-structured, actionable overview for an instruction-only skill: concrete script commands, a validated multi-step workflow with rollback, and clean progressive disclosure to real bundle files. Its only soft spots are a few non-essential narrative clauses and examples living in referenced files rather than inline.

Suggestions

Trim the bensz-collect-bugs collaboration convention and read-only integration notes to one line each, or move them to a references doc, to tighten conciseness toward 5.

Inline one short copy-paste example of the most common command (e.g. 'python scripts/run.py apply-section --project-root ... --title "国内外研究现状" --body-file ...') so the body alone is sufficient for the common case.

Dimension	Reasoning	Score
Conciseness	The body is lean with sectioned, non-padded guidance and no explanations of concepts Claude already knows (e.g. it does not explain what LaTeX or NSFC is). It is not a 5 because a few incidental narrative clauses (e.g. the bensz-collect-bugs collaboration convention and read-only integration notes) add tokens that are not strictly needed for the core writing task.	4 / 5
Actionability	Provides concrete executable guidance: exact script commands ('scripts/run.py coach --stage auto', 'scripts/run.py apply-section'), explicit write target ('extraTex/1.1.立项依据.tex'), and hard rules. It is not a 5 because copy-paste-ready code/examples live in scripts/README rather than inline; it is not a 3 because the inline commands are real and runnable, with only minor gaps.	4 / 5
Workflow Clarity	The 8-step recommended workflow is clearly sequenced with explicit validation checkpoints (引用可核验性守护, 与 2.1 研究内容的术语/缩写/指标一致性检查, 安全写入+备份, rollback/diff), and includes feedback loops (AI 不可用必须回退; 引用核验拒绝写入). This matches the 5 anchor for a destructive-write skill with validation.	5 / 5
Progressive Disclosure	The body is a clear overview with well-signaled one-level-deep references (references/.md, scripts/README.md, scripts/core/) that all resolve to real files, and content is appropriately split across references/scripts/assets rather than inlined. It matches the 5 anchor; navigation is easy and no nesting beyond one level is implied.	5 / 5
	Total	18 / 20 Passed

Description

87%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concrete, well-triggered, and tightly scoped to the NSFC 立项依据 niche, answering both what and when explicitly. Its only weakness is slightly less comprehensive action and synonym coverage than the top anchor.

Suggestions

Add 1-2 more concrete output-level actions (e.g. '生成可核验性诊断、术语一致性检查与安全写回') to lift specificity toward a 5.

Broaden trigger-term coverage with natural synonyms and file tokens (e.g. '标书', '申请书', '1.1.立项依据.tex', '.tex') for fuller keyword breadth.

Dimension	Reasoning	Score
Specificity	Names concrete outputs ('价值与必要性、现状不足、科学问题/假说与项目切入点') and an explicit constraint ('保持模板结构不被破坏'), giving several specific actions with only minor coverage gaps. It is not a 5 because the actions are described as deliverable categories rather than a comprehensive list of distinct concrete operations.	4 / 5
Completeness	Explicitly answers both 'what' (outputs the four justification sections and preserves template structure) and 'when' ('当用户明确要求...时使用') with concrete trigger phrases. It clearly matches the 5 anchor; stepping down to 4 would be unjustified since both what and when are stated explicitly and concretely.	5 / 5
Trigger Term Quality	Includes natural user-facing trigger phrases ('写/改 NSFC 立项依据', '立项依据写作/重构') plus relevant keywords (科学问题, 假说, 国内外现状), with only a few obvious synonyms/extension variants missing. It is not a 3 because coverage clearly exceeds 'some relevant keywords', and not a 5 because it lacks the full synonym/extension breadth (e.g. no file-type tokens like '.tex').	4 / 5
Distinctiveness Conflict Risk	It occupies a clear niche (NSFC/research-grant 立项依据 writing) with distinct, domain-specific triggers, creating minimal conflict risk with other skills. It matches the 5 anchor; the NSFC-specific framing makes overlap with generic skills unlikely.	5 / 5
	Total	18 / 20 Passed

Validation

75%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 deeper-than-1-level	Warning

	Total	12 / 16 Passed

Repository: huangwb8/ChineseResearchLaTeX
Path: skills/nsfc-justification-writer/SKILL.md
Commit: f9a7f2c

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.