make-latex-model

当用户明确要求“LaTeX 模板优化”“样式参数对齐”“像素级比对”“make-latex-model”或旧写法“make_latex_model”，或要把 ChineseResearchLaTeX 里的某个项目做成高质量模板时使用。适配 NSFC / paper / thesis / cv 四条产品线；先依据 packages/ 与 projects/ 的真实分层判断改项目层还是公共包，再用各产品线官方构建入口验收。若必须修改 packages 下公共包，需先生成受影响模板回归计划并完成相关回归；NSFC 专项工具仅在明确属于 NSFC 参数对齐场景时按需使用。

Quality

92%

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 content is a well-structured instruction-only skill: lean directives, a clearly sequenced workflow with validation and recovery, and proper one-level-deep reference splitting backed by real bundle files. The only weaknesses are minor boilerplate repetition and a few verification steps that name tools without invocation.

Suggestions

For the 官方入口验证 step, add the exact invocation (e.g. `python packages/bensz-nsfc/scripts/nsfc_project_tool.py build --project-dir <project>`) for each entry instead of only naming the tool, so the verification step is copy-paste ready.

Trim or externalize the 公共硬约束 / Skill 专属约束 boilerplate that overlaps with the canonical constraints shared across skills, to reclaim token budget for skill-specific guidance.

Add one concrete example invocation of a compare/analysis helper (e.g. compare_pdf_pixels.py with args) so the baseline-comparison path is as actionable as the regression gate.

Dimension	Reasoning	Score
Conciseness	The body is dense and directive with no padding of concepts Claude already knows, but the canonical 公共硬约束 block and some skill-specific constraints restate constraints that could be trimmed or delegated. Not a 5 because a few sections carry minor redundant boilerplate.	4 / 5
Actionability	Provides a concrete executable command for the package-regression gate (plan_package_regression.py with full path and argument) and names each product-line official entry script, but several verification steps only name tools without their invocation, leaving minor gaps. Not a 3 because the key gate is fully executable.	4 / 5
Workflow Clarity	The flow is clearly sequenced — 验收口径 → 修改层级 → 最小范围 → 包层安全门禁 (numbered with prove-first step) → 官方入口验证 → a 6-point 校验 checklist → 失败与恢复 — with an explicit validation checklist and a recovery-from-failure loop for the destructive package-layer change.	5 / 5
Progressive Disclosure	The body is an overview that clearly signals one-level-deep references (references/PRODUCT_LINE_RULES.md, references/SCRIPT_SCOPE.md, scripts/README.md, docs/*) which all exist as real bundle files, with the detailed product-line table and script scope correctly delegated to those files rather than inlined.	5 / 5
	Total	18 / 20 Passed

Description

100%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 explicit, action-oriented, and well-scoped: it tells Claude both what it does and exactly when to invoke it, with concrete user-quoted triggers and the old naming variant. It avoids vague fluff and over-claims entirely.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — judging project-layer vs common-package changes, validating via each product line's official build entry, generating a regression plan for affected templates, and pixel-level PDF comparison — giving comprehensive coverage across the four product lines.	5 / 5
Completeness	Explicitly states both what the skill does (land high-quality templates, pick the right layer, validate via official entries, run regression) and when to use it ("当用户明确要求…时使用") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Quotes natural trigger phrases users would actually say ("LaTeX 模板优化", "样式参数对齐", "像素级比对") plus the exact skill name and its old snake_case variant, covering synonyms and naming variants.	5 / 5
Distinctiveness Conflict Risk	Scoped to a clear niche (ChineseResearchLaTeX template alignment for NSFC/paper/thesis/cv) with distinct triggers and product-line-specific validation entries, minimizing overlap with unrelated skills.	5 / 5
	Total	20 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: huangwb8/ChineseResearchLaTeX
Path: skills/make-latex-model/SKILL.md
Commit: dd19421

Reviewed: 4 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.