DYZ MedPaper skill
32 skills for medical manuscripts in Claude Code and Codex: from topic choice and PubMed search to reviewer responses
- 369 ★ on the repository
- Last commit
Extended access
Why this level
- Texts and data from the manuscript folder go into the model session, so materials must be de-identified
- Local scripts read project files, query PubMed and write exported documents
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/dyz-medpaperDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Texts and data from the manuscript folder go into the model session, so materials must be de-identified
- Local scripts read project files, query PubMed and write exported documents
This is third-party code. Review the repository files before installing.
What it does
The set of 32 skills covers the path of a medical manuscript: idea and topic choice, literature search and review, data adequacy checks, statistics interpretation, introduction, methods, results, discussion, abstract, figures and tables, flowcharts, Chinese-English translation, journal selection, submission prep, reviewer replies and revision after a supervisor's comments. The dyz-med-orchestrator skill looks at the manuscript folder, names the most important bottleneck and picks the right skill. Work happens inside the project folder: the log, confirmed numbers and supervisor comments are written to files, so interrupted work can resume. There is an evidence audit, checks against guidelines such as PRISMA, systematic review templates, PubMed search through a built-in tool and export to a basic Word draft. The dyz-med-deai skill smooths templated language, does not hide AI use and is not positioned as detector evasion. The dialogue is in Chinese, and texts are written in Chinese and English.
Who it is for. Physicians and medical graduate students who write papers in Chinese and English and want an agent beside them from topic to reviewer response.
Good fit when
- A table holds patient data and it is unclear what paper it could become
- Results are computed but the discussion only repeats the numbers
- A supervisor returned the text noting weak novelty and a revision order is needed
- A long review arrived and the reply must be broken down point by point with edit locations found in the text
- A journal is being chosen for a manuscript within a publication-fee budget
Not a fit when
- Commercial use is planned: the PolyForm Noncommercial licence requires separate written permission
- You need clinical advice or a patient decision, the skills only prepare manuscripts
- Materials hold non-anonymized patient data, which must not be passed to a model
Example request
Use dyz-med-orchestrator: look at the manuscript folder, name the weakest spot and take the next stepLimitations
The project is in public beta: continuous work on a long manuscript and behaviour across agents are only lightly tested, and there is no independent review of the medical content. Docs and dialogue are mostly in Chinese. The PolyForm Noncommercial 1.0.0 licence allows non-commercial use only. The built-in search covers PubMed, while a multi-database review and Chinese databases need separate sources and access rights. The agent helps plan statistics, but they must run in an environment with the right software. Word export suits a basic review draft.
How to disable. Delete the 32 dyz-med-* folders from the agent's skills directory. Manuscripts live in the project folder and stay in place.
FAQ
How do I install DYZ MedPaper in Claude Code?
Clone the repository and run `python scripts/install_codex.py --target "~/.claude/skills"`, then the same command with `--check`. The main skill is invoked with `/dyz-med-orchestrator`.
Can the set be used in commercial projects?
The PolyForm Noncommercial 1.0.0 licence permits non-commercial use only. Anything else requires a separate written agreement described in COMMERCIAL-LICENSE.md.
Can the skill search PubMed?
Yes, a built-in tool searches PubMed. For Chinese databases such as CNKI the skill helps plan the search, but you need your own access.
What does the dyz-med-deai skill do?
It edits repetitive and templated language without changing facts. It does not evade AI detectors and does not remove disclosure of AI use.
How do I upgrade the installed skills?
Run the installer from the new version with `--upgrade --backup /path/to/new-backup`: it saves replaced files first, then verify with the `--check` command.
Discussion
No questions or reviews yet.
Related
Skillsscientific-writing-zh is a Claude Code skill, a Chinese-language writing coach for papers, revision, reviews and talks.
Skills for clinical research: literature search, statistics, reporting guideline and citation checks, journal submission
Skill with 30 prompt templates for research papers in English and Chinese, from literature review to grant proposals
Skills tailored to specific academic journals that help choose a venue, prepare a paper and pass review
A skill for Chinese-to-English paper translation: the author approves terms first, then it translates and reviews independently
A skill that searches five open databases for papers and builds an offline report with relevance, BibTeX and search strategies.
README in short
The Chinese README presents 32 skills for medical manuscripts, a table of where you are stuck and what you get, repository examples and installation for Claude Code and Codex. It states the public beta status, rules for patient data and the PolyForm Noncommercial licence. An English README and a docs site exist.
README badge
Original SKILL.md
--- name: dyz-med-orchestrator description: >- 读取医学论文项目状态与现有工件，生成当前进度地图、关键风险和一个最合适的下一步建议。当用户问“现在 到哪一步”“下一步做什么”“从哪里开始”或需要澄清多种医学任务的选择时使用；状态查询只推荐下一步， 明确的自然语言任务直接匹配最窄职责 Skill，无需说出技能名；不自动串联后续阶段或宣告稿件可投稿。 Use to review medical research workflow progress and recommend the next task. --- # 医学论文进度地图与路由 ## 定位 这是地图，不是自动流水线。它可以读取 `paper/project-passport.yaml` 和项目工件，解释当前状态并推荐下一步。 用户用自然语言提出的明确任务就是选择相应 Skill 的请求，无需手动说出技能名；仅问进度或下一步不构成执行授权。 ## 意图选择与授权边界 - 先匹配当前请求中职责最窄的 Skill，例如“只翻译这段摘要”直接选 `dyz-med-translate`，不先运行导航或摘要生成。 - “写讨论”“精读这三篇论文”等清晰请求可直接选对应 Skill；不要为确认技能名增加一轮提问。 - 只有请求确有歧义、且上下文不足以区分任务时才澄清，例如“帮我处理一下论文”需分清只读诊断、翻译或修订。 - “现在到哪一步”“下一步做什么”只读取状态并推荐，不执行推荐项，不自动把阶段串成整篇写作流水线。 - 选中 Skill 不等于其前置条件已满足。统计任务可先做只读检查和拟定计划，执行分析仍须核对针对当前计划的明确批准； 用户泛称“分析一下”或“继续”不能代替计划批准，也不能授权未列入计划的模型。 - 导出只生成本地成品，投稿包准备只生成本地材料。上传、发送、外部数据仓库创建和正式提交须另行明确授权并最终确认。 用户提及“投稿”可触发材料准备或授权澄清，但不得据此点击提交。 - 明确请求中所需的辅助工作仍受目标 Skill 自身边界约束；识别意图不授权所有后续阶段或自动循环修订。 ### 宿主前提 本库的 `SKILL.md` 描述提供自然语言触发条件，32 份 `agents/openai.yaml` 显式声明 `policy.allow_implicit_invocation: true`，供支持该字段的宿主兼容使用。Codex 按描述支持隐式选择， 该策略默认允许；这不是运行时分类器，也不证明 Breezell 或其他宿主读取此字段。实际调用仍依赖技能安装、 索引、宿主能力和会话工具权限。当前结构校验与路由用例不等于部署完成或真实宿主意图调用已验证。 ## 入口判断 先判断属于： - **新项目**：有想法、方案或数据，但没有完整稿件； - **现有稿件**：已有摘要、部分/完整草稿、拒稿信或审稿意见； - **状态查询**：项目已在进行，只需要进度、风险和下一动作。 只有身份或经验会实质改变建议时才询问，不把用户画像当作固定前置步骤。