ChineseResearchLaTeX is a collection of LaTeX templates and an AI-assisted workflow for preparing Chinese research documents such as grant proposals, papers, theses, and academic CVs. Researchers use it to plan, format, review, compile, and revise these documents with human oversight. The catalogue skills and instructions support its agent-based research-writing workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add huangwb8/ChineseResearchLaTeX --skill make-latex-modelgit clone --depth 1 https://github.com/huangwb8/ChineseResearchLaTeXWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/huangwb8/chineseresearchlatex/make-latex-model)<a href="https://agentmods.dev/skills/huangwb8/chineseresearchlatex/make-latex-model"><img src="https://agentmods.dev/badge/skills/huangwb8/chineseresearchlatex/make-latex-model/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/huangwb8/chineseresearchlatex/make-latex-model"><img src="https://agentmods.dev/badge/skills/huangwb8/chineseresearchlatex/make-latex-model.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00153 | $0.01742 |
| Opus 5 | $0.00077 | $0.00871 |
| Sonnet 5 | $0.00031 | $0.00348 |
| Haiku 4.5 | $0.00015 | $0.00174 |
Grade A, and why
make-latex-model scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ChineseResearchLaTeX 模板落地与高保真对齐器
- 让
ChineseResearchLaTeX中的目标项目按当前真实架构落成高质量模板。 - 先判断该改
projects/*还是packages/bensz-*。 - 若必须改公共包,先做回归计划,再跑受影响模板的官方验证。
- 验收始终以各产品线官方构建入口为准。
适用任务
- 把某个项目对齐到官方 PDF、Word 导出 PDF 或既有 baseline
- 判断问题属于项目层差异还是共享样式/共享脚本
- 做像素级 PDF 比对、标题对齐、参数抽取
- 新增或重构 NSFC / paper / thesis / cv 模板
流程
输入
按用户请求和配置文件提供必要输入;缺失信息应明确列出并停止依赖该输入的步骤。
执行步骤
-
当用户环境中出现因本 skill 设计缺陷导致的 bug 时,优先使用
bensz-collect-bugs按规范记录到~/.bensz-skills/bugs/,严禁直接修改用户本地 Claude Code / Codex 中已安装的 skill 源码。 -
若 AI 仍可通过 workaround 继续完成用户任务,应先记录 bug,再继续完成当前任务。
-
当用户明确要求“report bensz skills bugs”等公开上报动作时,调用本地
gh与bensz-collect-bugs,仅上传新增 bug 到huangwb8/bensz-bugs;不要 pull / clone 整个 bug 仓库。 -
产品线标准:
docs/for-developers/*-template-standard.md -
本 skill 工作流:
docs/WORKFLOW.md -
产品线识别:
references/PRODUCT_LINE_RULES.md -
脚本职责:
references/SCRIPT_SCOPE.md -
工具说明:
scripts/README.md -
基线准备:
docs/BASELINE_GUIDE.md
判断验收口径
- 用户要“像某份 PDF/Word 一样”
- 还是“按当前仓库标准做成好模板”
- 还是“新增一套模板能力”
判断修改层级
projects/*:示例内容、薄封装、项目资源、局部差异packages/bensz-*:共享样式、共享字体、profile、统一构建逻辑
最小范围实现
- 只改与当前任务直接相关的文件
- 除非用户明确要求,否则默认不改正文语义内容
包层安全门禁
当必须改 packages/ 时,额外执行:
- 先证明项目层方案不够
- 运行
python3 skills/make-latex-model/scripts/plan_package_regression.py <packages/bensz-*> - 优先把改动收敛到最窄的模板专属
profile/style/template - 改完先验目标项目,再回归该公共包直接覆盖的全部现有项目
官方入口验证
-
NSFC:
nsfc_project_tool.py -
Paper:
paper_project_tool.py -
Thesis:
thesis_project_tool.py -
CV:
cv_project_tool.py -
analyze_pdf.py -
compare_headings.py -
compare_pdf_pixels.py -
optimize_heading_linebreaks.py -
plan_package_regression.py
这些脚本是辅助工具箱,不是唯一工作流;NSFC 专项工具不能默认替代 paper / thesis / cv 的官方入口。
输出
输出 Skill description 所承诺的交付物,并明确格式、路径和失败返回形式。
输出管理
本 Skill 的新任务中间文件统一写入 ./.bensz-api/task-{yyyymmdd-hhmm}-{简短描述}/{skill名}/input|output|log/。同一任务复用一个任务根目录;多 Skill 协作才创建 shared/。正式交付物不写入该目录,历史隐藏目录只允许显式兼容读取、迁移或清理。
校验
- 改动落在正确层级
- 通过对应产品线官方入口
- warning 需要说明是已有还是新增
- 若改了公共包,必须说明回归了哪些模板
- 若用户给 baseline,完成必要 compare
paper默认兼顾 PDF 与 DOCX;cv默认兼顾中英文;thesis默认兼顾 profile/style 与项目入口一致性
What ships with it
59 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- CHANGELOG.md 3.3 KB
- config.yaml 9.8 KB
- docs/BASELINE_GUIDE.md 2.3 KB
- docs/FAQ.md 3.1 KB
- docs/WORKFLOW.md 4.8 KB
- output/README.md 245 B
- prompts/analysis_template.txt 1.1 KB
- README.md 5.6 KB
- references/PRODUCT_LINE_RULES.md 2.5 KB
- references/SCRIPT_SCOPE.md 1.7 KB
- scripts/__init__.py 36 B runs code
- scripts/analyze_pdf.py 9.2 KB runs code
- scripts/benchmark.bat 1.4 KB runs code
- scripts/benchmark.sh 2.5 KB runs code
- scripts/check_state.py 10 KB runs code
- scripts/compare_headings.py 48 KB runs code
- scripts/compare_paragraph_images.py 5.6 KB runs code
- scripts/compare_pdf_pixels.py 20 KB runs code
- scripts/convergence_detector.py 14 KB runs code
- scripts/core/__init__.py 181 B runs code
- scripts/core/ai_optimizer.py 5.4 KB runs code
- scripts/core/config_loader.py 6.2 KB runs code
- scripts/core/decision_reasoner.py 11 KB runs code
- scripts/core/diff_analyzer.py 9.1 KB runs code
- scripts/core/font_detector.py 7.0 KB runs code
- scripts/core/history_memory.py 1.9 KB runs code
- scripts/core/latex_format_parser.py 9.6 KB runs code
- scripts/core/paragraph_alignment.py 9.0 KB runs code
- scripts/core/parameter_executor.py 13 KB runs code
- scripts/core/template_base.py 7.5 KB runs code
- scripts/core/template_catalog.py 3.9 KB runs code
- scripts/core/validator_base.py 4.1 KB runs code
- scripts/core/validators/__init__.py 414 B runs code
- scripts/core/validators/compilation_validator.py 5.6 KB runs code
- scripts/core/validators/heading_validator.py 6.7 KB runs code
- scripts/core/validators/style_validator.py 7.1 KB runs code
- scripts/core/validators/visual_validator.py 5.0 KB runs code
- scripts/core/workspace_manager.py 17 KB runs code
- scripts/enhanced_optimize.py 29 KB runs code
- scripts/extract_headings_from_pdf.py 11 KB runs code
- scripts/extract_headings.py 9.7 KB runs code
- scripts/extract_paragraphs.py 2.7 KB runs code
- scripts/generate_baseline.py 17 KB runs code
- scripts/intelligent_adjust.py 22 KB runs code
- scripts/match_paragraphs.py 3.5 KB runs code
- scripts/optimize_heading_linebreaks.py 13 KB runs code
- scripts/optimize.bat 2.1 KB runs code
- scripts/optimize.py 12 KB runs code
- scripts/optimize.sh 2.9 KB runs code
- scripts/plan_package_regression.py 9.6 KB runs code
- scripts/prepare_main.py 9.4 KB runs code
- scripts/README.md 3.5 KB
- scripts/run_ai_optimizer.py 6.5 KB runs code
- scripts/run_validators.py 4.5 KB runs code
- scripts/setup_wizard.py 9.6 KB runs code
- scripts/sync_config.py 12 KB runs code
- scripts/validate.bat 5.0 KB runs code
- scripts/validate.sh 14 KB runs code
- workspace/NSFC_General/baseline/2026年最新word模板-1.面上项目-正文_analysis.json 748 B
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago Changed · -27 lines a33ee38cc5ad
- 12d ago First seen · 157 lines · 153 tokens per session scan A 010c95b071ec
make-latex-model is a skill published in the GitHub repository huangwb8/ChineseResearchLaTeX (2,726 stars, last pushed yesterday), licensed MIT. It adds 153 tokens to every session and 1,742 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
git-workflow
This skill should be used when the user asks to "create git commit", "manage branches", "follow git workflow", "use Conventional Commits", "handle merge conflicts", or asks about git branching strategies, version control best practices, pull request workflows. Provides comprehensive Git workflow guidance for team…
daily-paper-generator
Use when the user asks to generate daily paper digests on a general topic. This skill supports both arXiv and bioRxiv (or either one), then produces structured Chinese/English summaries for selected papers.
codex-autoresearch
Run autonomous, measurable experiments in a Git repository: change one hypothesis, verify a numeric metric, keep improvements, and revert failures. Use when the user wants Codex to keep iterating toward a numeric target in the foreground or as a detached background run. Do not use for ordinary one-shot coding…
clipboard
Copy text to clipboard with optional rich formatting. Triggers on "copy to clipboard", "copy that", "pbcopy", "copy formatted", "copy rich text".
neo4j-modeling-skill
Design, review, and refactor Neo4j graph data models. Use when choosing node labels vs relationship types vs properties, migrating relational/document schemas to graph, detecting anti-patterns (generic labels, supernodes, missing constraints), designing intermediate nodes for n-ary relationships, enforcing schema with…
git-commit
A Git assistant for reviewing local changes and creating conventional commit messages, which are standardized summaries of code changes. It can also run Git hooks and push the commit.