Medical Materials Figure Skill

v0.1.0

为医学材料、组织工程、骨修复、血管生成、支架机制图等论文配图生成稳定的分镜结构、风格约束与高质量提示词。Use when the user asks for biomaterials figures, scaffold schematics, bone regeneration, defect repair, a...

0· 89·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for chenxi-hub/medical-material-drawing.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Medical Materials Figure Skill" (chenxi-hub/medical-material-drawing) from ClawHub.
Skill page: https://clawhub.ai/chenxi-hub/medical-material-drawing
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install medical-material-drawing

ClawHub CLI

Package manager switcher

npx clawhub@latest install medical-material-drawing
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name and description (medical/biomaterials figure generation) align with the content of SKILL.md: detailed workflows, storyboard, style constraints, and prompt templates for generating publication-style biomedical schematics. The skill does not request unrelated binaries, env vars, or config paths.
Instruction Scope
SKILL.md stays within the stated scope: it extracts scientific narrative, proposes panel layouts, and produces core/style/negative prompts. It does mention 'call image generation tool' when the user explicitly requests a rendered image; no specific external endpoints are embedded. Note: because prompts may be sent to whatever image-generation service the agent uses, the operator should avoid including any patient-identifying or sensitive data in prompts.
Install Mechanism
Instruction-only skill with no install spec, no code files, and no downloads. Lowest-risk install profile; nothing is written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. There is no disproportionate access requested relative to the stated purpose.
Persistence & Privilege
The skill is not set to always:true and makes no requests to modify other skills or system configuration. Autonomous invocation is allowed by platform default but is not combined with elevated privileges or broad credential access.
Assessment
This skill appears coherent and lightweight, but consider these practical points before installing or using it: (1) It may instruct the agent to send prompts to your configured image-generation service — do not include any protected health information or identifying patient details in prompts. (2) Verify the image-generation provider's terms and copyright for publication use (journals sometimes require provenance or rights statements). (3) If you plan to publish figures, prefer generating editable SVGs or layered outputs as suggested so you can correct labels and ensure scientific accuracy. (4) The skill produces scientific schematics but does not verify factual claims — double-check any mechanism diagrams or asserted biology for correctness before publication. (5) If you integrate this skill with an external image API, review that integration's privacy/security settings so prompts and generated images are not leaked or stored in ways you don't expect.

Like a lobster shell, security has layers — review code before you run it.

latestvk977capcjse8wha2espr3j7z1984wbnd
89downloads
0stars
1versions
Updated 1w ago
v0.1.0
MIT-0

医学材料绘图 Skill

目标

产出适合论文、综述和图形摘要的医学材料示意图,默认追求以下效果:

  • 干净、专业、投稿级
  • 矢量感强,不走照片写实路线
  • 科学逻辑清楚,一眼能看懂“缺损/疾病 -> 材料设计 -> 修复结果”
  • 适合后续继续排版到 FigureTOC graphicgraphical abstract

何时使用

当用户提到以下任务时,优先使用这个 skill:

  • 医学材料论文配图
  • 骨缺损、牙槽嵴、软组织、皮肤、软骨、骨软骨修复示意图
  • 支架、微球、水凝胶、膜、涂层、复合材料、药物递送系统机制图
  • 血管生成、成骨、免疫调控、屏障作用、细胞迁移、组织再生流程图
  • “按之前那张图的风格继续做”
  • “做成论文图/图形摘要/机制示意图”

先判断用户要的交付物

  1. 如果用户只要提示词或构图方案,只输出提示词、分镜和版式建议,不主动生成图片。
  2. 如果用户明确要成图,再先整理结构化提示词,然后调用图像生成工具。
  3. 如果用户强调论文可编辑、后续还要改字改箭头,优先建议走 SVG 或可分层重绘路线,再视需要导出 PNG

默认工作流

按下面顺序做,不要一上来就直接堆提示词。

  1. 抽取科学主线
    • 缺损或病灶是什么
    • 材料由哪几层/哪几个模块组成
    • 每个模块承担什么功能
    • 生物学结果是什么
    • 时间顺序是否重要
  2. 选叙事框架
    • 最常用:问题/缺损 -> 材料设计 -> 修复结果
    • 如果强调机制:材料结构 -> 关键信号/细胞行为 -> 终点组织结果
    • 如果强调时序:implantation -> early response -> late regeneration
  3. 生成分镜草案
    • 默认横向 24 个 panel
    • 每个 panel 只承载一个核心信息
    • 每个 panel 给一个极短标题
  4. 再写提示词
    • 先写主体提示词
    • 再写风格约束
    • 最后补负面约束
  5. 生成后复盘
    • 是否像论文图,而不是商业插画
    • 是否标签拥挤
    • 是否科学关系被画反
    • 是否材料和宿主组织区分不清

默认版式

  • 优先白底
  • 优先横向排版
  • 默认 A/B/CA/B/C/D panel 字母放左上角
  • 每个 panel 顶部只保留一句短标题
  • 标签尽量放到结构外侧,用细引线连接
  • 箭头优先表示过程或方向,颜色不要滥用
  • 同一个对象在不同 panel 中保持造型一致

风格锚点

默认风格接近高质量医学示意图,而不是写实渲染图:

  • 解剖结构做适度简化,轮廓清晰
  • 边线细且稳,优先深灰或黑色,不要粗黑描边
  • 颜色柔和、低饱和、浅梯度
  • 允许轻微体积感,但不要金属质感或游戏建模质感
  • 材料结构要“可读”,例如通道、孔隙、分层、载药区要明确
  • 宿主组织、血管、新生骨、材料本体要一眼区分
  • 文本排版像期刊 figure,不像海报宣传图

推荐配色

可在不破坏风格统一的前提下替换,但默认先从这一套开始:

  • 软组织 / 牙龈:浅粉、肉粉、珊瑚粉
  • 骨组织 / 矿化基质:米黄、浅棕、骨色
  • 支架 / 水凝胶 / 功能材料:浅蓝、青蓝、灰蓝
  • 新生骨 / 成骨区域:青绿、蓝绿
  • 血管:暗红、砖红
  • 轮廓与引线:深灰,不用纯黑大面积压画面

强约束

下面这些约束要尽量固定,除非用户明确要求改:

  • 不要照片感皮肤、肉感、血腥渗出或真实手术场景
  • 不要高光塑料感、3D 游戏建模感、过强阴影
  • 不要背景纹理、花哨光效、漂浮装饰元素
  • 不要把标签压在关键结构上
  • 不要一张图里塞过多说明文字
  • 不要把不同层材料画得混成一团
  • 不要把箭头颜色用成彩虹

提示词写法

优先输出下面这三段,而不是一整坨长提示词。

1. Core prompt

先写清楚科学对象、场景和 panel 结构。

Create a clean publication-quality biomedical schematic illustration for a journal figure.
Theme: [disease/defect/material system].
Layout: [2-4]-panel horizontal figure labeled A, B, C[, D].
Panel A shows [problem / defect / baseline anatomy].
Panel B shows [material design / scaffold architecture / layered construct].
Panel C shows [early biological response / implantation / angiogenesis / immunomodulation].
Panel D shows [late regeneration / tissue repair / functional restoration].
Include clear external labels, thin leader lines, subtle arrows, white background, and consistent anatomy across panels.

2. Style prompt

把风格往“论文医学图”拉稳。

Style: high-end medical illustration, vector-like clarity, soft pastel palette, subtle gradients, precise outlines, simplified but anatomically credible tissues, clear material microstructure, elegant scientific figure design, not photorealistic, not poster-like, not cartoonish.

3. Negative prompt

Avoid photorealism, surgical gore, dramatic lighting, glossy 3D render style, cluttered labels, busy background, excessive text, inconsistent anatomy, oversaturated colors, fantasy design, comic style.

默认输出格式

如果用户没有指定,你可以先给出下面四项:

  1. 一句话图意概括
  2. 分镜草案
  3. 可直接复制的英文提示词
  4. 如需二次精修,再补中文改图指令

常见图型套路

缺损修复类

优先采用:

  • 缺损部位
  • 材料植入/结构设计
  • 早期反应
  • 晚期再生

多层材料类

优先强调:

  • 上层做什么
  • 中间层做什么
  • 下层做什么
  • 每层对应哪种生物功能

机制解释类

优先画清:

  • 材料释放什么
  • 作用于哪些细胞或组织
  • 导致哪些中间事件
  • 最终组织结局是什么

修图迭代规则

如果第一次生成不理想,优先按下列方向修,不要盲目重写全部提示词。

  • 结构不清:减少每个 panel 的信息量
  • 材料不像材料:强化 layered scaffold, radial channels, porous architecture, core-shell, mineralized matrix 之类的结构词
  • 解剖不稳:改成 simplified cross-sectional anatomy
  • 画面太花:强调 white background, minimal clutter, publication figure
  • 不像论文图:增加 journal figure, graphical abstract, scientific schematic, vector-like
  • 颜色脏或太艳:强调 soft pastel, low saturation, clean palette

回答模板

用户要你出方案时,优先按这个顺序回答:

  1. 图意
  2. 分镜建议
  3. 英文 prompt
  4. 负面约束
  5. 如需我直接生成/继续细化 SVG,请继续说

Additional resources

  • 示例见 [examples.md](examples.md)

Comments

Loading comments...