PDF Translate
Use when translating PDF documents between languages while strictly preserving original layout, formatting, typography, tables, and signatures. Powered by LLM direct comprehension, two-stage data-contract translation, chunked modular generation, in-browser JS overflow detection, and high-fidelity vector PDF reconstruction. Features zero-hallucination guardrails and automated 1:1 page-by-page audit script (audit_pdf.py).
Install
npx skills add lxsssssss/pdf-translate --skill pdf-translategit clone --depth 1 https://github.com/lxsssssss/pdf-translate.git /tmp/pdf-translate
mkdir -p ~/.claude/skills
cp -r /tmp/pdf-translate/. ~/.claude/skills/pdf-translatePersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# PDF 保持原排版翻译 Skill (大模型直译 + 几何探针测量 + 矢量排版重构 + 生成后对比自愈) 当用户需要**翻译 PDF 文档**,且要求**必须完整保留原始版面格式、公文表头、目录、双语表格、签章栏、插图物理尺寸**并最终**输出为高保真矢量 PDF 文件**时,使用此 Skill。 > [!IMPORTANT] > **四大核心铁律(杜绝机翻乱码、长文档幻觉、跨页溢出与图片排版变形)**: > 1. **严禁脑补,严格忠实直译 (Strict Faithful Translation & Zero Hallucination)**: > - **不得压缩或遗漏条款**:逐条忠实翻译每一个一级、二级、三级子条款(如法定资质证照清单 2.1.1~2.1.8、年限指标、年产能门槛 150,000 吨、分卷装订规则 Volume 2A/2B 等),坚决禁止主观总结或合并条款; > - **坚决禁止虚构常识**:严禁凭常识脑补原文不存在的交付程序、装订份数、银行授信金额或办事流程(例如原文未写具体金额则绝对禁止编造 1,000,000 KWD,无 5.3 节则绝对不可添加 5.3 节); > - **表格格式 1:1 忠实还原**:官方附表(如 Appendix-1 至 Appendix-6)的列名、行高、选项框、备注及签字盖章区必须严格对齐原件,严禁用自创的“申报核对清单 (Checklist)”或通用表格粗糙代替。 > 2. **几何探针与视觉锚点测量铁律 (Geometry Probe & Visual Anchor Measurement)**: > - **绝对禁止经验主义猜度尺寸**:严禁在未测量原件几何信息的情况下,随意给插图或表格加上 `max-height: 220px` 等主观限制,导致原图严重缩水变形或比例失调; > - **原图物理坐标与尺寸 1:1 提取**:提取图片时必须调用 `page.get_image_rects(xref)` 获取原图在页面上的实际 `width`、`height` 以及 `x0, y0`,并在 HTML 中严格以真实物理尺寸呈现(如原件宽 218.9pt、高 322.4pt,HTML 中必须按照该物理尺寸渲染); > - **几何线条与装饰框高保真复刻**:通过 `page.get_drawings()` 提取所有几何元素(如顶会论文标题框的 4pt 粗黑顶线与 1pt 细底线、长度 143.5pt 的左对齐短横脚注线),严禁用通用单薄的 100% 宽度边框粗暴替代; > - **中文信息密度补偿与垂直基线对齐**:中文字符信息密度通常比西文高 20%~35%,必须记录关键文字块的垂直基准起点 `y0`(如摘要标题、正文、章节起始与脚注线位置),通过微调行高(`line-height: 1.35~1.45`)、段落间距(`margin`)以及灵活运用网格与绝对定位,确保中文版面的视觉重心与垂直基线与原件严密锚定,杜绝页面下半部分突兀大面积留白; > - **首页页眉页脚特殊规则**:大部分顶会论文(NIPS/ICML/CVPR 等)或官方公文首页(Page 1)不编排页码,底部通常为会议发表信息或官方备案行,严禁在第一页机械添加页码。 > 3. **两阶段解耦与分块生成 (Schema-First & Chunked Pipeline)**:
Details
- Skill name
pdf-translate- Repository
- lxsssssss/pdf-translate
- License
- MIT
- Repo updated
- 23 October 2026
FAQ
What does the PDF Translate skill do?
Use when translating PDF documents between languages while strictly preserving original layout, formatting, typography, tables, and signatures. Powered by LLM direct comprehension, two-stage data-contract translation, chunked modular generation, in-browser JS overflow detection, and high-fidelity vector PDF reconstruction. Features zero-hallucination guardrails and automated 1:1 page-by-page audit script (audit_pdf.py).
How do I install the PDF Translate skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add lxsssssss/pdf-translate --skill pdf-translate. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is PDF Translate popular?
It lives in lxsssssss/pdf-translate, which has 93 GitHub stars and 1 forks.
Related skills
Many Ppt Skills
67brycewang-stanford/many-ppt-skills
Pick an AI slide-deck skill and a concrete visual style from a curated registry, filtering on the requirements that decide it — editable in PowerPoint, speaker notes, a mandated corporate template, offline, PDF — with sample imagery and the style ids each project actually uses. Use when the user wants to make a presentation, deck or slides and has not already chosen a tool; asks which slide skill to use or what the difference between them is; wants to know what a style looks like before committing; or names a style id such as soft-editorial or swiss-grid. This skill routes to the skill that ma
Pakco HTML
76pakco77/pakco-html
Creates HTML decks, webpages, and social cards with pakco.html templates. Use for slides, PPT-style reports, visual web pages, or the style picker.
PDF Proof
77metedata/pdf-proof
Use this skill when the user needs visual proof that specific values exist in a PDF — not just to read a PDF, but to see exactly where a number, amount, clause, or field appears on the page with a highlighted screenshot. Trigger when someone is cross-referencing a PDF against something else (a form they're filling out, a claim, a conversation, another document) and needs confirmation with evidence. Common signals: entering data into TurboTax or a form and pulling source numbers from last year's return; asking an accountant to verify reported figures; checking contract terms before sending to l
Report
87v-yun/vuln-report-skill
确认漏洞后产出正式 SRC/0day 提交稿 DOCX 报告的收口 skill。仅在漏洞已确认、准备成稿时使用——负责查重、过分层验证门、按固定 DOCX 骨架生成报告(Heading 2 章节 + Step 式 PoC + 内嵌真实截图)、语义化命名、放对目录、处理驳回追加。不负责挖掘/验证漏洞本身。TRIGGER:用户说"写报告/出报告/成稿/提交稿/生成漏洞报告",或漏洞已验证到位准备交付时;或 /report。EN: Turns confirmed vulnerabilities into submission-ready DOCX reports for SRC/0day platforms (verification gates, fixed layout, step-style PoC, mandatory real screenshots). Trigger on "write report" or /report.
Indian Health Insurance Claim Help
60DhananjayBhosale/Indian-Health-Insurance-Claim-Help
Prepare Indian health-insurance (mediclaim) reimbursement claims for hospital, day-care and pre/post costs. Lists needed documents, finds missing bills, reports and receipts, builds upload PDFs.
Resumify
97breaker505/resumify
Transform any internship experience into resume-ready, interview-ready expressions — for all roles: dev, product, operations, design, sales, finance.
Browse the most popular Claude skills, all documents & office files skills, or write your own with the skills generator.