菜单
PDF / 图片 / 网页高质量解析为 Markdown。三种部署形态全部免费：本地 CLI（开源）、 云 API（注册免费 token + 配额，超额才付费）、网页 flash（零注册，单文件 ≤10MB / ≤20页）。 在中文学术 PDF / 表格 / 公式上识别质量显著高于 pdfplumber。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Step 5 of the PaperOrchestra pipeline (arXiv:2604.05018). Iteratively refine drafts/paper.tex by simulating peer review and applying targeted revisions, with strict accept/revert halt rules. Maintains a worklog and snapshots each iteration so revert is real, not symbolic. TRIGGER when the orchestrator delegates Step 5 or when the user asks to "refine the draft", "iterate on the paper", or "run peer review on this paper".
Step 3 of the PaperOrchestra pipeline (arXiv:2604.05018). Execute the literature search strategy from outline.json — discover candidate papers via web search, verify them through Semantic Scholar (Levenshtein > 70 fuzzy title match, temporal cutoff, dedup by paperId), build a BibTeX file, and draft Introduction + Related Work using ≥90% of the verified pool. Runs in parallel with the plotting-agent. TRIGGER when the orchestrator delegates Step 3 or when the user asks to "find citations for my paper", "draft the related work", or "build the bibliography".
Step 1 of the PaperOrchestra pipeline (arXiv:2604.05018). Convert (idea.md, experimental_log.md, template.tex, conference_guidelines.md) into a strict JSON outline containing a plotting plan, literature search plan (Intro + Related Work), and section-level writing plan with citation hints. TRIGGER when the orchestrator delegates Step 1 or when the user asks to "outline a paper from raw materials" or "generate the paper structure".
Orchestrate the full PaperOrchestra (Song et al., 2026, arXiv:2604.05018) five-agent pipeline to turn unstructured research materials (idea, experimental log, LaTeX template, conference guidelines, optional figures) into a submission-ready LaTeX manuscript and compiled PDF. TRIGGER when the user asks to "write a paper from my experiments", "turn this idea and these results into a paper", "generate a conference submission", "run paper-orchestra on X", or otherwise wants the end-to-end paper-writing pipeline. Coordinates the outline-agent, plotting-agent, literature-review-agent, section-writing-agent, and content-refinement-agent skills.
Use when solving CUMCM, MCM, ICM, or other mathematical modeling tasks that need contest problem intake, model selection, Python solving, figures, paper writing, quality audit, and packaged deliverables.
Use when EZ_math_model needs to convert paper.md to paper.docx, read a DOCX problem statement, preserve equations and tables, or package a modeling report as a Word document.
| name | mineru |
| description | PDF / 图片 / 网页高质量解析为 Markdown。三种部署形态全部免费：本地 CLI（开源）、 云 API（注册免费 token + 配额，超额才付费）、网页 flash（零注册，单文件 ≤10MB / ≤20页）。 在中文学术 PDF / 表格 / 公式上识别质量显著高于 pdfplumber。 |
| allowed-tools | Bash, Read |
| read_when | ["题目 PDF 含表格 / 公式 / 复杂排版","阅读 zhanwen 优秀论文 PDF","把附件中带数据的报告 PDF 转为 markdown"] |
external/zhanwen-mathmodel/ 下的优秀论文本子 skill 不自带 Python 实现；调用宿主已安装的 mineru-open-api CLI
或 mineru Python 包。详细 CLI 文档见 mineru 官方 skill。
文件 ≤ 10MB 且 ≤ 20页 且 不需表格识别
→ flash-extract（零配置、最快）
文件 > 10MB 或 > 20页 或 需要表格 / 公式 / OCR
→ extract（要 token，注册免费领）
→ 备用：本地 CLI mineru（开源，自动下载模型）
# Flash 模式（推荐起手）
mineru-open-api flash-extract problem.pdf -o workdir/.../out --language ch
# 限页（>20 页时）
mineru-open-api flash-extract problem.pdf -o workdir/.../out --language ch --pages 1-20
# Extract 精确模式（需 token；写入 EZMM_MINERU_TOKEN）
mineru-open-api auth # 一次性配置
mineru-open-api extract problem.pdf -o workdir/.../out --language ch --table --formula
# 本地 CLI（无 token，开源版）
pip install -U mineru
mineru -p problem.pdf -o workdir/.../out
| 参数 | 默认 | 说明 |
|---|---|---|
--language | ch | 中文文档；英文用 en；混合 ch_server |
--pages | 全部 | flash 模式下题目 > 20 页时按 1-20 分批 |
-o | stdout | 写入指定目录（推荐）；不写参数则 markdown 走 stdout |
| Env Var | 必需 | 用途 |
|---|---|---|
EZMM_MINERU_TOKEN | extract 模式必需 | 注册免费领，flash 不需要 |
| 错误 | 原因 | 处理 |
|---|---|---|
file page count exceeds API limit | flash 限 20 页 | 加 --pages 1-20；或换 extract（要 token） |
TLS handshake timeout | 网络抖动 | 重试 1 次；仍失败降级 pdfplumber |
no API token found | extract 缺 token | 调 mineru-open-api auth 或 fallback flash |
输出 Markdown 含大量 <!-- image --> | 论文图片占位 | 正常；只对文字部分有效 |