安装方式
手动下载安装
下载 ZIP 后解压到技能目录即可安装。若在桌面客户端 WebView中直接下载出现异常,本站会改为提示页 + 原始链接,请按页内说明操作。
下载 ZIP (shub-pandoc-convert-openclaw-v1.0.0.zip)触发指令
/pandoc-convert
跨平台安装指引
该技能声明兼容以下 1 个平台,将 ZIP 解压到对应目录即可被识别。
unzip shub-pandoc-convert-openclaw-v1.0.0.zip -d ~/.claude/skills/
mkdir -p 创建;启用 Skill 后请重启对应 Agent 让配置生效。
使用指南
Pandoc 转换(OpenClaw)
围绕 Pandoc 转换(OpenClaw):在 OpenClaw 工作流中串联 Pandoc 做文档互转;模板与引擎参数见 SKILL.md。 无需在每次任务前把零散英文说明手工拼进上下文,也 减少 与客户端默认行为脱节的试错;具体命令、钩子与 JSON 参数仍以 ZIP 包内 SKILL.md 为权威。下文结构与站内 MCP CLI 类专题稿相同:何时用、前置、流程、速查与故障。
何时使用
- 在 OpenClaw 工作流中串联 Pandoc 做文档互转
- 模板与引擎参数见 SKILL.md
- 已获取本技能 ZIP,并准备在 Claude Code / OpenClaw 中按 SKILL.md 挂载。
- 希望用中文专题稿快速判断「该不该启用」,再深入英文 SKILL 查参数与边界。
- 需要与团队对齐同一套触发方式、目录约定或回调格式时。
前置条件
- 通用:可运行 Claude Code 或文档要求的客户端;有可读写的项目工作区(或 SKILL.md 指定的沙箱目录)。
- 权威细节:API Key / OAuth、钩子路径、环境变量以 ZIP 内 SKILL.md 为准。
- OpenClaw:实例可访问本机或远端的 Claude Code / 脚本路径,网络与权限与当前部署一致。
典型流程
- 从 ClawHub / 站内分发获取技能 ZIP,校验版本与校验和(若提供)。
- 阅读 SKILL.md 的安装段落:目录落点、客户端类型(Claude Code / OpenClaw / 脚本)。
- 用文档中的最小示例完成第一次调用(单文件修改、单次查询或单次委派)。
- 确认工作目录、权限边界与输出路径后,再处理多文件或长耗时任务。
- 需要回调 / Webhook / 通知时,按 SKILL.md 配置端点并在测试环境先验通。
与 ZIP / SKILL.md 的关系
站内专题稿与 MCP CLI 类 oss 稿同样:概括何时用、怎么接、怎么排错;命令模板、钩子名、JSON 字段、版本矩阵一律以 ZIP 内 SKILL.md 与 ClawHub 上游为准。
命令示例(摘自包内 SKILL.md)
以下为从上游 SKILL.md(或入库正文)自动抽取的终端/脚本片段;路径、环境变量与参数以当前 ZIP 与官方说明为准。
ClawHub slug:pandoc-convert-openclaw(安装命令以 SKILL.md / claw CLI 为准)。
# Single file conversion
python scripts/convert.py input.md output.pdf
# With custom template
python scripts/convert.py report.md report.pdf --template business --toc
# Batch convert
python scripts/convert.py --batch *.md --format pdf --output-dir ./pdfs
# Batch convert with validation
./scripts/batch_convert.sh input/*.md pdf output/
# Validate output
./scripts/validate.sh output/document.pdf
./scripts/validate.sh output/book.epub
# Markdown → PDF
pandoc input.md -o output.pdf
# Markdown → Word
pandoc input.md -o output.docx
# Word → Markdown
pandoc input.docx -o output.md --extract-media=./media
python scripts/convert.py [OPTIONS] INPUT OUTPUT
Options:
--format FORMAT Force output format
--template TEMPLATE Use named template
--toc Include table of contents
--number-sections Number headings
--title TITLE Document title
--author AUTHOR Document author
--batch Batch mode
--validate Validate output
--verbose Detailed output
./scripts/batch_convert.sh INPUT_DIR FORMAT OUTPUT_DIR [OPTIONS]
# Example
./scripts/batch_convert.sh ./docs/ pdf ./output/ --toc --number-sections
./scripts/validate.sh FILE
# Validates:
# - PDF structure and readability
# - EPUB spec compliance (requires epubcheck)
# - HTML validity
# - File integrity
站内入库时的触发命令(完整语义见 ZIP):
# 使用本技能时可在对话中引用或执行上述指令;完整参数与示例见下载包内 SKILL.md。
/pandoc-convert
最佳实践
- 先 SKILL.md 再猜参数;站内专题稿不替代 schema 与必填字段说明。
- 委派任务时写清验收标准(命令、文件路径、测试命令),减少来回追问。
- 长任务用文档推荐的回调 / 日志落盘代替高频轮询,省 Token 也省机器负载。
- 多技能同时启用时,注意钩子加载顺序与重复工具调用(以 SKILL.md 冲突说明为准)。
调试与排错
- 打开 stderr 与客户端日志;PTY/tmux 场景同时看面板最后几十行输出。
- 参数错误时对照 SKILL.md 中的 JSON/CLI 示例(引号、转义、工作目录)。
- 网络类失败:查代理、防火墙、MCP 传输方式(stdio / HTTP / SSE)。
- OpenClaw 侧核对 channel、路由与 Claude Code 进程是否仍在运行。
速查
| 动作 | 说明 |
|------|------|
| 获取技能包 | ClawHub / 站内 ZIP,核对版本 |
| 权威步骤 | 优先阅读 ZIP 内 SKILL.md |
| 首次试跑 | 使用 SKILL.md 最小示例 |
| 验收 | 对照路径、测试命令或回调负载 |
| OpenClaw | 路由、channel 与远端进程状态 |
常见故障
- 无输出或立即退出 → 工作目录错误、依赖未装、或 Claude Code 未登录;按 SKILL.md 自检清单执行。
- 权限被拒绝 → 检查沙箱路径、
--permission-mode与工具白名单。 - 与简介不符 → 以英文 SKILL 与上游仓库为准,站内稿仅作结构化导读。
# 📄 Pandoc Convert (Integrated)
**Universal document converter combining unified Python tools with modular bash utilities.**
The **pandoc-convert** skill provides intelligent workflows for converting documents between 40+ formats using pandoc. This integrated version combines:
- **Unified Python converter** (convert.py) - Single powerful tool for most conversions
- **Modular bash utilities** (batch_convert.sh, validate.sh) - Specialized workflows
- **Comprehensive templates** - Both LaTeX academic and modern CSS styles
- **Professional documentation** - Complete guides, troubleshooting, and references
## ✨ Key Features
- **40+ Format Support**: Markdown, Word, PDF, HTML, LaTeX, EPUB, RST, AsciiDoc, Org-mode, and more
- **Dual Toolset**: Python for smart conversions + bash for validation/batch processing
- **Professional Templates**: 12 templates covering academic, business, and web use cases
- **Comprehensive Documentation**: Format guides, troubleshooting, templates, and quick reference
- **Smart Defaults**: Optimized settings for each conversion path
- **Metadata Preservation**: Keep titles, authors, dates across formats
- **Error Recovery**: Validation and helpful error messages
## 🔧 Prerequisites
### Required
- **pandoc** (v2.19+ recommended)
- **Python 3.8+** (for convert.py helper)
### Optional (for extended formats)
- **LaTeX** (TeX Live, MiKTeX) - Required for PDF output
- **wkhtmltopdf** - Alternative HTML to PDF converter
- **librsvg** - SVG support
- **epubcheck** - EPUB validation
See `INSTALL.md` for detailed installation instructions per platform.
## 📚 Quick Start
### Using Python Helper (Recommended)
```bash
# Single file conversion
python scripts/convert.py input.md output.pdf
# With custom template
python scripts/convert.py report.md report.pdf --template business --toc
# Batch convert
python scripts/convert.py --batch *.md --format pdf --output-dir ./pdfs
```
### Using Bash Utilities
```bash
# Batch convert with validation
./scripts/batch_convert.sh input/*.md pdf output/
# Validate output
./scripts/validate.sh output/document.pdf
./scripts/validate.sh output/book.epub
```
### Direct Pandoc
```bash
# Markdown → PDF
pandoc input.md -o output.pdf
# Markdown → Word
pandoc input.md -o output.docx
# Word → Markdown
pandoc input.docx -o output.md --extract-media=./media
```
## 🎯 Common Workflows
See `references/conversion-guides.md` for detailed step-by-step guides:
- Markdown → Professional PDF (business reports, academic papers)
- Word → Markdown (version control friendly)
- Markdown → EPUB (eBooks with validation)
- Multi-file → Single PDF (book compilation)
- Markdown → HTML5 (standalone with CSS)
## 🎨 Templates
### LaTeX Templates (Academic/Professional)
- `academic-paper.tex` - Manuscript style
- `business-letter.tex` - Professional correspondence
- `technical-report.tex` - Technical documentation
- `resume.tex` - CV/resume formatting
- `professional.tex` - General-purpose professional
- `report-template.tex` - Report structure
### CSS Templates (Web/Modern)
- `github.css` - GitHub markdown style
- `blog-style.css` - Clean blog format
- `epub-style.css` - eBook styling
- `presentation.html` - HTML presentations
- `ebook.css` - Enhanced eBook layout
### Reference Documents
- `reference-styles.docx` - Word style reference
All templates in `templates/` directory.
## 🔧 Tool Reference
### convert.py (Python)
Unified conversion tool with smart defaults:
```bash
python scripts/convert.py [OPTIONS] INPUT OUTPUT
Options:
--format FORMAT Force output format
--template TEMPLATE Use named template
--toc Include table of contents
--number-sections Number headings
--title TITLE Document title
--author AUTHOR Document author
--batch Batch mode
--validate Validate output
--verbose Detailed output
```
### batch_convert.sh (Bash)
Batch processing with progress tracking:
```bash
./scripts/batch_convert.sh INPUT_DIR FORMAT OUTPUT_DIR [OPTIONS]
# Example
./scripts/batch_convert.sh ./docs/ pdf ./output/ --toc --number-sections
```
### validate.sh (Bash)
Post-conversion validation:
```bash
./scripts/validate.sh FILE
# Validates:
# - PDF structure and readability
# - EPUB spec compliance (requires epubcheck)
# - HTML validity
# - File integrity
```
## 📊 Format Support
### Input Formats
**Markdown**: `markdown`, `gfm`, `markdown_mmd`
**Word**: `docx`, `odt`, `rtf`
**Web**: `html`, `html5`
**LaTeX**: `latex`, `tex`
**Plain Text**: `txt`, `rst`, `textile`, `asciidoc`
**Academic**: `jats`, `docbook`
**Presentation**: `pptx`
**eBooks**: `epub`
**Other**: `json`, `csv`, `org`, `mediawiki`, `man`
### Output Formats
All input formats plus: **PDF**, **EPUB**, **RevealJS**, **Beamer**
Complete format matrix: `references/format-matrix.md`
## 🗂️ Directory Structure
```
pandoc-convert-integrated/
├── SKILL.md # This file
├── INSTALL.md # Detailed installation guide
├── README.md # Quick start guide
├── scripts/
│ ├── convert.py # Unified Python converter
│ ├── batch_convert.sh # Bash batch processor
│ └── validate.sh # Validation utility
├── templates/
│ ├── *.tex # LaTeX templates (6)
│ ├── *.css # CSS templates (3)
│ ├── *.html # HTML templates (1)
│ └── *.docx # Word reference (1)
└── references/
├── format-guide.md # Format details
├── format-matrix.md # Compatibility matrix
├── conversion-guides.md # Step-by-step guides
├── format-support.md # Supported features
├── quick-reference.md # Cheat sheet
├── templates.md # Template documentation
└── troubleshooting.md # Problem solving
```
## 🐛 Troubleshooting
### Common Issues
- **"pandoc: command not found"** → Install pandoc (see INSTALL.md)
- **"pdflatex not found"** → Install LaTeX distribution
- **Unicode broken in PDF** → Use `--pdf-engine=xelatex`
- **Images missing** → Check paths and use `--resource-path`
- **EPUB validation fails** → Run epubcheck for details
See `references/troubleshooting.md` for comprehensive solutions.
## 📖 References
- `INSTALL.md` - Platform-specific installation
- `references/format-guide.md` - Format capabilities and limitations
- `references/conversion-guides.md` - Step-by-step workflows
- `references/quick-reference.md` - One-page cheat sheet
- `references/templates.md` - Template usage and customization
- `references/troubleshooting.md` - Extended problem solving
## 🎯 Best Practices
1. **Use YAML frontmatter** for metadata (title, author, date)
2. **Validate outputs** before sharing (especially EPUB/PDF)
3. **Version control source** (Markdown), not outputs
4. **Test templates first** before batch processing
5. **Back up before batch operations**
## 🚀 Performance
- Use `batch_convert.sh` for parallel processing of multiple files
- Cache templates in `~/.pandoc/templates/`
- Use incremental builds (only reconvert changed files)
- For very large docs (>10MB), increase memory limits
## 📜 License
This skill is part of OpenClaw. Pandoc itself is GPL-licensed.
---
**Quick Start**: `python scripts/convert.py input.md output.pdf`
**Batch Convert**: `./scripts/batch_convert.sh *.md pdf ./output/`
**Validate**: `./scripts/validate.sh output.pdf`
**Help**: See `README.md` and `references/` directory