Tdoc DOCX

What it does

When your agent uses it

• Tasks that involve Word documents

Example prompts

• “/tdoc-docx”

Requirements

• Python 3
• Node.js
• A Bash shell

What it can do on your machine

Read from SKILL.md and the folder at commit e5199b5. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 13 files in scripts/ (Python and Shell, from the files we listed), which the agent can run.

  Shell commands in SKILL.md call:

  • python3
  • apt
  • brew
  • bash
  • pip3
  • uv
  • npm
  • soffice

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use pip3, uv and npm, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Tdoc DOCX loads about 3.6k tokens when it runs. Until then it costs about 76 tokens; SKILL.md has 676 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~76

When it runs · the whole SKILL.md, loaded when a task matches

~3.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NoteRuns commands with sudoSKILL.md:113

  w install --cask libreoffice`<br>Linux: `sudo apt install libreoffice` |

• NoteRuns commands with sudoSKILL.md:114

  macOS: `brew install pandoc`<br>Linux: `sudo apt install pandoc` |

• NoteRuns commands with sudoSKILL.md:115

  macOS: `brew install poppler`<br>Linux: `sudo apt install poppler-utils` |

• NoteRuns commands with sudoSKILL.md:116

  acOS: `brew install antiword`<br>Linux: `sudo apt install antiword` |

• NoteRuns commands with sudoSKILL.md:619

  - Linux: `sudo apt install fonts-noto-cjk`

• NoteRuns commands with sudoSKILL.md:625

  - Linux: `sudo apt install libreoffice`

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from LeoYeAI/openclaw-master-skills at commit e5199b5, republished under its MIT licence (© LeoYeAI). 676 words, ~3,565 tokens.

name

tdoc-docx

description

Word 文档全能处理技能 | Complete Word Document Processing Skill. 支持创建、读取、编辑、转换 Word 文档 | Create, read, edit, convert Word documents. 支持 .docx/.doc 格式、中文公文格式、表格、图片、tracked changes、评论 | Supports .docx/.doc, Chinese gov format, tables, images, tracked changes, comments. 触发词：Word、文档、docx、doc、公文、报告、转PDF.

version

1.0.0

用户请求
  │
  ├─ 「创建」新文档 → 走路由决策树（路径 A 通用 / 路径 B 垂类模板）
  │
  └─ 对「已有」文档进行操作 → 直接使用 scripts/ 工具脚本
      ├─ 编辑/修改  → edit_docx.py（JSON 规则）或 XML 编辑（office/unpack → 修改 → pack）
      ├─ 读取/提取  → read_docx.py
      ├─ 格式转换   → convert_docx.py
      ├─ 差异对比   → diff_docx.py
      ├─ 评论/修订  → comment.py + XML 编辑
      └─ 字数/分析  → word_count.py + AI

cd {baseDir}
./install.sh

# 使用 pip
pip3 install -r {baseDir}/requirements.txt

# 或使用 uv（更快）
uv pip install -r {baseDir}/requirements.txt

用户请求创建文档
  │
  ├─ Step 1: 意图识别 — 是否匹配垂类模板？
  │   │
  │   ├─ ✅ 匹配垂类模板（公文、合同等专业文档）
  │   │   └─ → 路径 B：垂类模板创建流程
  │   │       ① 读取 {baseDir}/templates/<模板名>/rules.md
  │   │       ② 按 rules.md 规范，用 Python 脚本创建文档
  │   │       ③ 不使用 create_docx.py 的内置 style
  │   │
  │   └─ ❌ 不匹配垂类模板（通用文档）
  │       └─ → 路径 A：通用创建流程
  │           │
  │           ├─ Step 2: 用户是否上传/提供了 Markdown 文件？
  │           │   │
  │           │   ├─ ✅ 有 Markdown → 路径 A1：CLI 方式（--from-markdown）
  │           │   └─ ❌ 无 Markdown → 路径 A2：Python API 方式（⭐ 默认推荐）
  │           │
  │           └─ 选择风格: default / business / academic
  │
  └─ 输出文档

import sys
sys.path.insert(0, "{baseDir}/scripts")
from create_docx import DocxCreator

creator = DocxCreator(style='default')  # 可选: default/business/academic
creator.add_title("文档标题")
creator.add_heading1("一、第一章")
creator.add_paragraph("正文内容")
creator.add_paragraph("详细说明...", bold_prefix="（一）小标题。")
creator.add_table(["列1", "列2"], [["A", "B"], ["C", "D"]])
creator.add_image("chart.png", width=400, caption="图1")
creator.add_empty_line()
creator.add_page_break()
creator.save("output.docx")

# 从 Markdown 转换（自动识别标题层级）
python3 {baseDir}/scripts/create_docx.py --from-markdown input.md --output output.docx

# 带署名
python3 {baseDir}/scripts/create_docx.py --from-markdown input.md --output output.docx \
  --author "某某单位\n2026年3月11日" --style default

# 指定模板风格
python3 {baseDir}/scripts/create_docx.py --from-markdown input.md --output output.docx --style business

# 文档大标题

## 一、一级标题

### （一）二级标题

正文段落内容。

**1. **带加粗前缀的段落

- 列表项会转为段落

读取 {baseDir}/templates/<模板名>/rules.md

# 基本文本提取
python3 {baseDir}/scripts/read_docx.py document.docx

# 指定输出格式
python3 {baseDir}/scripts/read_docx.py document.docx --format json
python3 {baseDir}/scripts/read_docx.py document.docx --format markdown
python3 {baseDir}/scripts/read_docx.py document.docx --format text

# 提取特定内容
python3 {baseDir}/scripts/read_docx.py document.docx --extract text
python3 {baseDir}/scripts/read_docx.py document.docx --extract tables
python3 {baseDir}/scripts/read_docx.py document.docx --extract metadata
python3 {baseDir}/scripts/read_docx.py document.docx --extract images

# 批量处理
python3 {baseDir}/scripts/read_docx.py ./docs_folder --batch --format json --output results.json

# 输出到文件
python3 {baseDir}/scripts/read_docx.py document.docx --format markdown --output output.md

python3 {baseDir}/scripts/edit_docx.py input.docx output.docx edits.json

{
  "description": "修改说明",
  "replacements": [
    {
      "search": "旧文本",
      "replace": "新文本",
      "style": "highlight"
    }
  ],
  "additions": [
    {
      "after": "在此文本之后",
      "text": "添加的文本",
      "style": "bold"
    }
  ]
}

python3 {baseDir}/scripts/office/unpack.py document.docx unpacked/

<!-- 插入 -->
<w:ins w:id="1" w:author="Claude" w:date="2026-03-11T00:00:00Z">
  <w:r><w:t>插入的文本</w:t></w:r>
</w:ins>

<!-- 删除 -->
<w:del w:id="2" w:author="Claude" w:date="2026-03-11T00:00:00Z">
  <w:r><w:delText>删除的文本</w:delText></w:r>
</w:del>

<!-- 最小化编辑: 仅标记变化部分 -->
<w:r><w:t>期限为</w:t></w:r>
<w:del w:id="1" w:author="Claude" w:date="...">
  <w:r><w:delText>30</w:delText></w:r>
</w:del>
<w:ins w:id="2" w:author="Claude" w:date="...">
  <w:r><w:t>60</w:t></w:r>
</w:ins>
<w:r><w:t>天。</w:t></w:r>

python3 {baseDir}/scripts/comment.py unpacked/ 0 "评论内容"
python3 {baseDir}/scripts/comment.py unpacked/ 1 "回复内容" --parent 0

python3 {baseDir}/scripts/office/pack.py unpacked/ output.docx --original document.docx

python3 {baseDir}/scripts/accept_changes.py input.docx output.docx

# DOCX 转 PDF
python3 {baseDir}/scripts/convert_docx.py input.docx --to pdf --output output.pdf

# DOCX 转 Markdown
python3 {baseDir}/scripts/convert_docx.py input.docx --to markdown --output output.md

# DOC 转 DOCX（需要 LibreOffice）
python3 {baseDir}/scripts/convert_docx.py input.doc --to docx --output output.docx

# DOCX 转图片（需要 LibreOffice + Poppler）
python3 {baseDir}/scripts/convert_docx.py input.docx --to images --output ./pages/

# 生成 Unified Diff 报告
python3 {baseDir}/scripts/diff_docx.py old.docx new.docx --output diff_report.md

# 输出到终端
python3 {baseDir}/scripts/diff_docx.py old.docx new.docx

# 从上传获取最新文件
bash {baseDir}/scripts/fetch_file.sh upload output.docx

# 从本地路径
bash {baseDir}/scripts/fetch_file.sh ~/Documents/report.docx output.docx

# 从 URL 下载
bash {baseDir}/scripts/fetch_file.sh https://example.com/file.docx output.docx

# 从 SFTP
bash {baseDir}/scripts/fetch_file.sh sftp://user@host:/path/file.docx output.docx

# 统计文档字数（输出到终端）
python3 {baseDir}/scripts/word_count.py document.docx

# 输出为 JSON 格式（方便程序处理）
python3 {baseDir}/scripts/word_count.py document.docx --format json

# 同时输出前 N 字预览
python3 {baseDir}/scripts/word_count.py document.docx --preview 200

<w:t>这里&#x2019;是引号：&#x201C;你好&#x201D;</w:t>

<!-- 评论标记是 w:p 的直接子元素，不在 w:r 内部 -->
<w:commentRangeStart w:id="0"/>
<w:r><w:t>被评论的文本</w:t></w:r>
<w:commentRangeEnd w:id="0"/>
<w:r><w:rPr><w:rStyle w:val="CommentReference"/></w:rPr><w:commentReference w:id="0"/></w:r>

cd {baseDir} && ./install.sh

© LeoYeAI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 41 other files (scripts) in skills/tdoc-docx of LeoYeAI/openclaw-master-skills.

• SKILL.md
• _meta.json
• install.sh
• pyproject.toml
• requirements.txt
• scripts/__init__.py
• scripts/accept_changes.py
• scripts/comment.py
• scripts/convert_docx.py
• scripts/create_docx.py
• scripts/diff_docx.py
• scripts/edit_docx.py
• scripts/fetch_file.sh
• scripts/office/__init__.py
• scripts/office/helpers/__init__.py
• scripts/office/helpers/merge_runs.py
• scripts/office/helpers/simplify_redlines.py
• scripts/office/pack.py
• … and 24 more

Open the folder on GitHubat commit e5199b5