Parse PDFs or arXiv IDs into structured markdown notes with figure analysis.
npx skills add https://github.com/Gonglitian/agent-skills --skill read-paper-gonglitian
Please help me install this Agent Skill. Skill: read-paper Source: https://github.com/Gonglitian/agent-skills/tree/main/skills/read-paper Command: npx skills add https://github.com/Gonglitian/agent-skills --skill read-paper-gonglitian
Read and summarize research papers (PDFs or arXiv IDs) into a structured markdown note that includes text summaries and figure analysis, saving time and enabling quick review.
Provide a paper PDF path or arXiv ID and I will generate a structured markdown note with figure analyses.
High-intent search queries and answers about installing and using this skill.
To generate structured notes from an arXiv paper, provide the arXiv ID to the Skill. It parses the PDF, extracts text and figures via VLM, and outputs a comprehensive markdown note with embedded figures.
Yes, you can extract and analyze figures from a PDF automatically using the integrated VLM figure-description step. The pipeline extracts images, generates descriptions, and cross-validates them in a figure_descriptions.json file.
To summarize long research papers, the Skill deploys dynamic parallel subagents that divide the document into chunks. This parallel-processing workflow extracts text and figures efficiently before assembling the final markdown note.
Yes, this paper-reading tool works with local PDF files by accepting a direct PDF path as input. It uses pymupdf to extract text and images, enabling you to create academic notes without needing an arXiv ID.
The best way to create academic notes with embedded figures is using a VLM-powered deep reading pipeline. It parses the document, analyzes the figures, and assembles everything into a structured note.md file for quick review.
A limitation of using VLM for figure analysis is that the generated descriptions require cross-validation to ensure consistency. The pipeline addresses this by outputting a figure_descriptions.json file to verify accuracy.
acertainKnightLocate and download academic PDFs from DOIs, arXiv IDs, or titles.
allenhutchisonConvert Obsidian notes into queryable bases with configurable views and filters.
unfoldingWordCoordinate splitting, parallel tn-writer execution, and merging of issue TSVs into notes TSV.
u9401066Extract text, figures, and citations from PDFs into Markdown.
