agentsclimarketplace

Pdf

Skill woodfishhhh/EZ_math_model/skills/ez-math-model/tools/pdf

专攻数学建模的下一代自动化SKIILS

Install
npx -y skills add woodfishhhh/EZ_math_model --skill pdf

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Use when EZ_math_model needs to read or convert PDF problem statements, extract text or tables from PDF attachments, summarize reference papers, or fall back when MinerU is unavailable.

SKILL.md

1.8 KB, as published. Nobody here has run it

pdf — PDF 读取与转换兜底

何时使用

  • 用户上传 .pdf 题目。
  • 需要读取优秀论文 PDF 的摘要、目录或局部章节。
  • MinerU 不可用或输出失败。
  • PDF 扫描件需要 OCR。

优先链路

  1. mineru skill。
  2. 宿主 pdf skill。
  3. pdfplumber 文本和表格提取。
  4. 图像 OCR 兜底。

论文 PDF 导出兜底

scripts/runtime/export_paper.ps1 的 simple text-only PDF 只用于占位交付。该兜底 使用基础字体,非 ASCII 字符会被替换为 ?,因此中文 formal 论文不能把它视为 高保真 PDF。出现该兜底时必须:

  • export_report.json 中写 pdf_fallbackpaper_pdf_high_fidelity=false
  • export_audit.json 中标记 pdf_readability=placeholder 或 warning;
  • 最终质量等级最高为 provisional_pass

题目 PDF 转 Markdown

import pdfplumber

with pdfplumber.open("problem.pdf") as pdf:
    text = "\n\n".join(page.extract_text() or "" for page in pdf.pages)

with open("problem.md", "w", encoding="utf-8") as f:
    f.write(text)

复杂表格、公式和多栏排版优先交给 mineru

参考论文读取策略

  • 优先读标题、摘要、目录、参考文献。
  • 只在需要借鉴模型表述时读对应章节。
  • 每篇累计读取页数不超过 5 页。
  • 不全文复制优秀论文内容。

失败诊断

情况处理
全部 PDF 工具不可用保留原 PDF,让多模态能力直接读
提取乱码写诊断,建议用户提供文本版
公式提取失败保留原始字符串,由 writer 补 LaTeX

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.