nsfc-reviewers

当用户明确要求"评审国自然标书"、"模拟专家评审"、"审阅 NSFC 申请书"时使用。模拟领域专家视角对 NSFC 标书进行多维度评审，输出分级问题与可执行修改建议。⚠️ 不适用：用户只是想写/改标书某个章节（应使用 nsfc-*-writer 系列技能）、只是想了解评审标准（应直接回答）、没有明确"评审/审阅"意图。

Quality

80%

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

78%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.

A well-structured, action-oriented workflow doc with clear sequencing and excellent progressive disclosure through verified one-level-deep references. Minor gains are available by trimming infrastructural boilerplate and adding an explicit validation feedback loop for the batch parallel-review step.

Suggestions

Add an explicit validate→fix→retry checkpoint for the parallel multi-panel review step (e.g., what to do when a panel output is malformed) to strengthen workflow_clarity toward 5.

Move or compress the BenszAPI task-workspace and bensz-collect-bugs collaboration sections into a short pointer or separate reference, trimming non-review boilerplate to improve conciseness.

Convert the parallel-vibe invocation and aggregation steps into concrete copy-pasteable commands where possible to raise actionability.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence (no library/format primers), but infrastructural boilerplate sections (BenszAPI task workspace, bensz-collect-bugs collaboration) add tokens that are not pure review guidance.	4 / 5
Actionability	Provides a concrete executable script invocation and references real scripts/config keys with explicit decision rules (ceil(N * consensus_threshold), P0→P1→P2), but the parallel-vibe and aggregation steps are procedural rather than copy-paste commands.	4 / 5
Workflow Clarity	Clear 7-step sequence with explicit checkpoints and graceful degradation ('数量为 0 时直接失败', '允许个别 thread 缺失但不能中断'), though the batch parallel review lacks an explicit validate→fix→retry feedback loop.	4 / 5
Progressive Disclosure	A dedicated '关键脚本与参考' section points one-level-deep to real, verified files (references/expert_.md, aggregation_rules.md, master_prompt_template.md, scripts/.py), with the body serving as a well-signaled overview.	5 / 5
	Total	17 / 20 Passed

Description

82%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.

A strong description that clearly states both what the skill does and when to use it, with concrete trigger phrases and explicit boundary guidance against related writer skills. The main weakness is specificity: it names only a couple of concrete actions rather than a comprehensive list.

Suggestions

Add one or two more concrete actions (e.g., '生成函评/会评阶段判断与翻盘动作', '合并跨组共识问题并按 P0/P1/P2 排序') to lift specificity from 3 toward 5.

Include a couple of common synonyms users might say (e.g., '改标书意见', '函评模拟') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain (NSFC 标书多维度评审) and two concrete actions ('模拟领域专家视角...多维度评审', '输出分级问题与可执行修改建议'), but does not list several specific actions for comprehensive coverage.	3 / 5
Completeness	Explicitly answers both what (multi-dimensional expert review with graded problems and actionable suggestions) and when (当用户明确要求...时使用) with concrete trigger phrases, plus explicit non-applicability guidance.	5 / 5
Trigger Term Quality	Includes several natural trigger phrases users would say ('评审国自然标书', '模拟专家评审', '审阅 NSFC 申请书') with good coverage, but is missing common synonyms/variations, so it does not reach comprehensive level 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (NSFC proposal expert review) with explicit anti-conflict guidance distinguishing it from nsfc-*-writer skills and from general Q&A, giving minimal conflict risk.	5 / 5
	Total	17 / 20 Passed

Validation

81%

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 — 13 / 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
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: huangwb8/ChineseResearchLaTeX
Path: skills/nsfc-reviewers/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.