Install
openclaw skills install paper-readerComprehensive PDF paper reader for academic research. Extracts text, figures, tables, and structured content from research papers with support for multimodal analysis.
openclaw skills install paper-reader专为学术论文阅读设计的多功能PDF提取工具。
pip install pdfplumber pymupdf Pillow
# 完整提取论文(文本+图片+表格)
python3 ~/.openclaw/skills/paper-reader/read_paper.py paper.pdf --full
# 仅提取文本
python3 ~/.openclaw/skills/paper-reader/read_paper.py paper.pdf --text
# 提取图片到指定文件夹
python3 ~/.openclaw/skills/paper-reader/read_paper.py paper.pdf --images --img-dir ./figures
# 提取表格为CSV
python3 ~/.openclaw/skills/paper-reader/read_paper.py paper.pdf --tables --csv-dir ./tables
# 提取前3页(通常包含摘要和引言)
python3 ~/.openclaw/skills/paper-reader/read_paper.py paper.pdf --pages 1-3
from paper_reader import PaperReader
reader = PaperReader("paper.pdf")
# 获取元数据
metadata = reader.get_metadata()
# 提取全文
text = reader.extract_text()
# 提取特定章节
abstract = reader.extract_section("Abstract")
methods = reader.extract_section("Methods")
# 提取图片
figures = reader.extract_images(output_dir="./figures")
# 提取表格
tables = reader.extract_tables()
paper_output/
├── paper.txt # 完整文本
├── metadata.json # 论文元数据
├── abstract.txt # 摘要
├── sections/ # 分章节文本
│ ├── 01_Introduction.txt
│ ├── 02_Related_Work.txt
│ ├── 03_Methods.txt
│ ├── 04_Experiments.txt
│ └── 05_Conclusion.txt
├── figures/ # 提取的图片
│ ├── fig_1.png
│ ├── fig_2_table.png
│ └── fig_3_chart.png
└── tables/ # 提取的表格
├── table_1.csv
└── table_2.csv
自动识别并提取:
识别Methods章节中的:
提取参考文献列表,支持BibTeX格式导出。