paper-parse
Parse academic PDF papers into markdown with figure extraction.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/chen-li-17/paper-parse-figuresWhat This Skill Does
The paper-parse skill is a specialized utility designed for researchers, students, and academics who frequently interact with scholarly PDF documents. By leveraging the power of PyMuPDF and pymupdf4llm, this agent skill automates the tedious process of converting rigid PDF layouts into structured, readable Markdown files. Beyond simple text extraction, the tool identifies and isolates figures and captions, creating a dedicated repository of visual assets. It essentially bridges the gap between static academic publishing formats and modern LLM-compatible text formats, allowing you to ingest complex research papers into your AI workflows without losing structural integrity.
Installation
To integrate this skill into your OpenClaw environment, use the following CLI command provided by the ClawHub registry:
clawhub install openclaw/skills/skills/chen-li-17/paper-parse-figures
This command automatically resolves the necessary dependencies, including PyMuPDF and the pymupdf4llm library, through uv. Ensure you have uv installed in your path for a seamless setup experience.
Use Cases
This skill is ideal for several professional scenarios:
- Literature Reviews: Quickly convert a folder of fifty PDF papers into Markdown to build a local vector database for Retrieval-Augmented Generation (RAG).
- Academic Summarization: Extract text and figures from new pre-prints to generate automated briefing documents or summaries.
- Knowledge Management: Convert dense research documents into Obsidian or Notion-compatible Markdown files, complete with referenced images.
- Data Extraction: Programmatically pull figure-based datasets from papers for further analysis.
Example Prompts
- "OpenClaw, please parse the PDF located at ~/downloads/transformer_paper.pdf and place the markdown output into my research folder."
- "Extract all figures from the paper in ./input/quantum-research.pdf and organize them into the /figures folder for my upcoming presentation."
- "Convert the academic paper at ./data/diffusion-models.pdf to markdown and identify the authors in the parsed metadata json."
Tips & Limitations
For optimal results, ensure the PDF files are not scanned images (OCR-based PDFs might require additional preprocessing). The skill works best with standard academic two-column layouts. If the output directory is not specified, it defaults to the script's local context, so it is recommended to always define an --output-dir to keep your project workspace clean. Note that highly complex figures with embedded sub-figures may occasionally require manual oversight to ensure captions are correctly mapped to their respective images.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-chen-li-17-paper-parse-figures": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write, code-execution
Related Skills
paper-card-analyzer
Analyze `paper-parse` outputs and generate a research-oriented paper card directly in natural language. Use this skill after paper parsing when you need a structured summary of contributions, method, experiments, limitations, reproducibility notes, and future work without running any extra script.
xhs-post-factory
小红书多输入内容生成技能。用于将 pdf/md/txt/json 等文件转为结构化的小红书博文。默认生成论文解读（paper-interpretation）类型，输出 xhs-post.md 与 xhs-post.json 到输入文件所在目录，并保留可扩展模板机制以支持后续更多博文类型。
xiaohongshu-operate
小红书运营执行技能。用于按给定素材执行发帖草稿创建（普通图文/长文/视频）、评论区检查与回复流程、关键词搜索并抓取已有内容。适用于用户明确要求“直接按素材操作平台”而不是重新生成内容的场景，默认发帖到草稿箱、评论先汇总后确认、抓取输出为 Top 20 的标准字段 Markdown。