$ loading_
You have a batch of paper DOIs or PDFs and need to pull out result tables, parameter tables, or comparison tables quickly. Instead of copying data manually paper by paper, you want a workflow that processes them in bulk and outputs a spreadsheet ready for further analysis.
这套方案按“拿到论文 PDF → 提取表格 → 汇总成表”的顺序组织，适合处理一批学术论文中的表格数据。agents365-ai-paper-fetch 负责根据 DOI 批量找到论文 PDF 来源，vlln-pdf-figures-mcp-server 专门从学术 PDF 中抽取图表与表格数据，最后用 composiohq-document-skills-xlsx 把多篇论文的结果整理进统一的数据表，方便后续清洗、比对与分析。
└ Use it first to find and retrieve research PDF download links from multiple sources based on DOI. This prepares the source files needed for the next table extraction step.
└ Feed the academic PDFs collected in the previous step into this tool to programmatically extract table data. The output becomes structured intermediate data for the final consolidation step.
└ Write and organize the extracted table outputs from multiple papers into a unified XLSX spreadsheet, making the data ready for filtering, comparison, statistics, and export.
Use `agents365-ai-paper-fetch` first to obtain PDF download links from DOIs, then run `vlln-pdf-figures-mcp-server` to extract table data from academic PDFs, and finally use `composiohq-document-skills-xlsx` to write all extracted results into one spreadsheet. This follows the dependency order of source first, extraction second, and output last.
If you already have all the PDFs locally, `agents365-ai-paper-fetch` is not strictly required. Its main value is finding and retrieving PDF download links from multiple sources based on DOI. With files already in hand, you can start directly from the table extraction step.
It is best suited for extracting figures and table data from academic PDFs, especially research papers. If your goal is to structure table content for downstream analysis, this is the core step. Its focus is chart and table extraction rather than general PDF editing.
The most common issue is inconsistent column names and column order across papers. `composiohq-document-skills-xlsx` is well suited for consolidating results into a spreadsheet, but you should define a unified schema before writing data so multiple papers can fit into one master table for filtering and analysis.