Extract clinical trial data from pharmaceutical conference websites or PDF documents. Use when user provides a URL or PDF file containing innovative drug clinical trial data and needs structured extraction of: drug name, manufacturer, indication, clinical phase, trial name, conference, efficacy and safety data (presented as tables), and markdown output to "药品名称@适应症.md" file.
This skill enables extracting structured clinical trial data from pharmaceutical conference websites (ASCO, ESMO, EHA, etc.) and saving it as a markdown file with standardized format.
输出路径: ~/.openclaw/workspace
命名格式: {药品名称}@{适应症}.md
文件名清理规则:
常见终点缩写列表 - 以下缩写在表格中无需写中文全称：
⚠️ 如需修改配置，请直接编辑本配置区域
⚠️ 注意：~ 需在执行时展开为实际用户 home directory
⚠️ 注意：不在列表中的终点数据应写中文全称以清晰说明
⚠️ 注意：本技能会尝试自动提取网页图片，但对于受限平台（微信公众号等）需要手动截图
This skill requires the following tools to be available in the OpenClaw runtime environment:
browser action=start profile=openclaw target=hostFor browser automation to work correctly:
/usr/bin/google-chrome or similar path)The following tools are used for PDF extraction. The skill will attempt each method in order:
nano-pdf CLI (recommended)
pdftotext (poppler-utils)
sudo apt-get install poppler-utilsbrew install poppler~/.openclaw/workspace)All configuration options are defined in the Configuration section above and can be modified without reinstalling the skill:
Use this skill when:
Examples of trigger phrases:
Determine if user provided a URL or a PDF file.
Use the built-in browser to open and extract page content:
Start browser (if not already running):
browser action=start profile=openclaw target=host
Navigate to URL:
browser action=open targetUrl=<provided-url>
Capture page snapshot to extract content:
browser action=snapshot format=markdown
Optional: Take screenshot for visual reference:
browser action=screenshot fullPage=true
Extract text content from the PDF. Two approaches available:
Approach 1: Use nano-pdf CLI (read-only)
nano-pdf --file <path-to-pdf> --action read
Approach 2: Use nano-pdf with natural language instructions
nano-pdf --file <path-to-pdf> --action edit --instruction "Extract all text content from this PDF, focusing on clinical trial data including drug name, indication, phase, efficacy, and safety results"
Note: The extracted PDF content will be in raw text format. You may need to clean up formatting before proceeding to extraction.
Analyze the fetched content and extract the following fields. Leave blank if information is not available:
网页图片：
图片处理的两种方式：
方式一：自动提取（适用于公开网站）
方式二：手动截图（适用于受限平台或提取失败时）
## 临床数据图片
⚠️ 无法自动提取图片（受限平台或提取失败），建议手动截图保存。
参考图片描述：
1. 疗效数据图（如 rPFS 曲线、OS 曲线）
2. 安全性汇总表（AE 发生率、严重 AE）
截图保存路径示例：
```markdown
PDF 图片：
For effectiveness and safety data, present findings in markdown table format:
## 药品有效性和安全性
| 指标 | ABC001 | 对照组 | HR | p-value |
|------|----------------|--------|------|------|
| N | 100 | 50 | - | - |
| ORR | 41.4% | 25.3% | - | <0.0001 |
| cORR | 34.5% | - | - | <0.0001 |
| DCR | 87.9% | - | - | <0.0001 |
| mPFS | 11.3 | 6.8 | 0.62 | <0.0001 |
| mOS | 22.1 | 14.2 | 0.73 | <0.0001 |
| 最常见AE | 恶心、血液事件（1-2级） | - | - | - |
多剂量组示例：
| 指标 | AAB001 2mg | AAB001 4mg | AAB001 6mg | Placebo |
|------|----------|--------------|--------------|--------------|
| N | 50 | 50 | 50 | 50 |
| OS | 12.1 | 14.2 | 17.3 | 0.2 |
| OS p-value | <0.0001 | <0.0001 | <0.0001 | - |
| PFS | 12.1 | 14.2 | 17.3 | 0.2 |
| PFS p-value | <0.0001 | <0.0001 | <0.0001 | - |
表格格式规范：
AAB001 (最大效果) 或 高剂量组AAB001 6mg 或 对照组["指标", "实验组1", "实验组2", ...] 或 ["指标", "实验组", "对照组"]（如有对照）11.3 而非 11.3个月）41.4%）N/A 或 NE（未成熟/未评估）表示11.3 (N=82)）Generate output file using the configuration from Configuration section:
{filename}, then expand ~ to actual home directory.From a medical/pharmaceutical expert perspective, provide a concise analysis of the clinical trial data. This section should be clearly marked as "（仅供参考）" (For reference only).
Key aspects to analyze:
Efficacy Evaluation
Safety Considerations
Study Design Assessment
Clinical Prospects
Cautions & Limitations
Provide concise, objective analysis (3-6 bullet points). Avoid over-optimistic language.
The generated markdown file should follow this template:
# {药品名称} - {适应症} 临床数据
## 基本信息
| 字段 | 内容 |
|------|------|
| 药品名称 | {药品名称} |
| 生产厂家 | {生产厂家} |
| 适应症 | {适应症} |
| 临床阶段 | {临床阶段} |
| 临床名称 | {临床名称} |
| 学术会议 | {学术会议} |
## 药品有效性和安全性
| 指标 | {实验组名称} |
|------|--------------|
| 主要终点数据... | 值 |
| 次要终点数据... | 值 |
| 安全性数据... | 值 |
## 试验设计
| 设计要素 | 内容 |
|----------|------|
| 研究类型 | ... |
| 入组人数 | ... |
## 临床数据图片
{网页图片链接或PDF截图引用}
## 专家点评
（仅供参考）
从药学/医学专家角度分析该临床数据的意义：
- **疗效评价**：[分析主要终点结果是否达到临床意义，对比同类药物]
- **安全性考量**：[分析安全性概况，关注关键AE]
- **研究设计评价**：[研究设计是否合理、样本量是否充足、对照组选择等]
- **临床前景**：[基于当前数据评估药物商业化潜力及后续研究方向]
- **注意事项**：[数据的局限性、需要进一步验证的点等]
## 数据来源
{URL或PDF路径}
提取时间: {当前日期}
输出路径 设置target=host 参数。如果浏览器未运行，skill 会自动启动 或 <URL>（避免大图预览）PD-1抑制剂_图表1.png--action read 快速提取全文内容Not every skill requires all three types of resources.