📖 简介

MCP Word Commander 是一个基于 MCP (Model Context Protocol) 的 Word 文档处理服务，允许 AI 助手直接读取、编辑 Word 文档。

✨ 核心功能

📦 依赖

• Python 3.10+
• Docker（可选，用于容器化部署）
• 详见 requirements.txt

🚀 快速开始

方式一：本地运行

# Windows
.venv\Scripts\activate
pip install -r requirements.txt
python server.py

# macOS / Linux
source venv/bin/activate
pip install -r requirements.txt
python server.py

方式二：Docker 部署

🐳 使用 Docker Compose（推荐）

# 构建并启动
docker-compose up -d --build

# 查看日志
docker-compose logs -f

# 停止服务
docker-compose down

🐳 使用 Docker 命令

# 构建镜像
docker build -t mcp-word-commander:latest .

# 运行容器（挂载本地 documents 目录）
docker run -it --rm \
  -v $(pwd)/documents:/documents \
  mcp-word-commander:latest

# Windows PowerShell
docker run -it --rm `
  -v ${PWD}/documents:/documents `
  mcp-word-commander:latest

方式三：在 Claude Desktop / Cursor 中配置

{
  "mcpServers": {
    "word-commander": {
      "command": "python",
      "args": ["D:/Program/mcp_word_edit/server.py"],
      "env": {}
    }
  }
}

{
  "mcpServers": {
    "word-commander": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-v", "D:/Documents:/documents",
        "mcp-word-commander:latest"
      ],
      "env": {}
    }
  }
}

🛠️ 功能列表

📄 文档基础操作

📝 段落操作

🖼️ 图片操作

📊 表格操作

💡 使用示例

🖼️ 读取文档中的图片

# AI 可以直接"看到"文档中的图片
read_images("document.docx")

# 只读取第一张图片
read_images("document.docx", image_index=0)

📸 在指定位置插入图片

# 在第 3 段后插入图片，设置宽度为 4 英寸，居中对齐
insert_image_after_paragraph(
    file_path="document.docx",
    after_index=2,
    image_path="image.png",
    width_inches=4.0,
    alignment="CENTER"
)

📊 在指定位置插入表格

# 在第 5 段后插入 3x3 表格
insert_table_after_paragraph(
    file_path="document.docx",
    after_index=4,
    rows=3,
    cols=3,
    data=[
        ["姓名", "年龄", "城市"],
        ["张三", "25", "北京"],
        ["李四", "30", "上海"]
    ],
    header_bold=True
)

✏️ 修改表格单元格

# 修改第一个表格的 (1, 2) 单元格
update_table_cell(
    file_path="document.docx",
    table_index=0,
    row=1,
    col=2,
    new_text="新内容",
    font_name="Microsoft YaHei",
    font_size=12,
    is_bold=True
)

📁 项目结构

mcp_word_edit/
├── 📄 server.py           # MCP 服务器主文件
├── 📋 requirements.txt    # Python 依赖
├── 🐳 Dockerfile          # Docker 镜像配置
├── 🐳 docker-compose.yml  # Docker Compose 配置
├── 🚫 .dockerignore       # Docker 构建忽略文件
├── 📖 README.md           # 说明文档
└── 📂 documents/           # 文档目录（Docker 挂载点）

🐳 Docker 相关

💾 数据持久化

使用 Docker 时，建议将本地目录挂载到容器的 /documents 目录：

docker run -it --rm -v /path/to/your/docs:/documents mcp-word-commander:latest

📦 镜像信息

⚡ 技术特点

🤝 贡献

欢迎提交 Issue 或 Pull Request！

在提交前，请确保：

• ✅ 代码风格一致
• ✅ 包含必要的说明文档
• ✅ 通过相关测试

📜 许可证

本项目使用 MIT 许可证，详见 LICENSE 文件。

📖 Overview

MCP Word Commander is a Word document processing service based on MCP (Model Context Protocol), enabling AI assistants to directly read and edit Word documents.

✨ Core Features

📦 Requirements

• Python 3.10+
• Docker (optional, for containerized deployment)
• See requirements.txt for details

🚀 Quick Start

Option 1: Local Installation

# Windows
.venv\Scripts\activate
pip install -r requirements.txt
python server.py

# macOS / Linux
source venv/bin/activate
pip install -r requirements.txt
python server.py

Option 2: Docker Deployment

🐳 Using Docker Compose (Recommended)

# Build and start
docker-compose up -d --build

# View logs
docker-compose logs -f

# Stop service
docker-compose down

🐳 Using Docker Command

# Build image
docker build -t mcp-word-commander:latest .

# Run container (mount local documents directory)
docker run -it --rm \
  -v $(pwd)/documents:/documents \
  mcp-word-commander:latest

# Windows PowerShell
docker run -it --rm `
  -v ${PWD}/documents:/documents `
  mcp-word-commander:latest

Option 3: Configure in Claude Desktop / Cursor

{
  "mcpServers": {
    "word-commander": {
      "command": "python",
      "args": ["/path/to/mcp_word_edit/server.py"],
      "env": {}
    }
  }
}

{
  "mcpServers": {
    "word-commander": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-v", "/path/to/documents:/documents",
        "mcp-word-commander:latest"
      ],
      "env": {}
    }
  }
}

🛠️ Feature List

📄 Basic Document Operations

📝 Paragraph Operations

🖼️ Image Operations

📊 Table Operations

💡 Usage Examples

🖼️ Read Images from Document

# AI can directly "see" images in the document
read_images("document.docx")

# Read only the first image
read_images("document.docx", image_index=0)

📸 Insert Image at Specific Position

# Insert image after paragraph 3, set width to 4 inches, center alignment
insert_image_after_paragraph(
    file_path="document.docx",
    after_index=2,
    image_path="image.png",
    width_inches=4.0,
    alignment="CENTER"
)

📊 Insert Table at Specific Position

# Insert a 3x3 table after paragraph 5
insert_table_after_paragraph(
    file_path="document.docx",
    after_index=4,
    rows=3,
    cols=3,
    data=[
        ["Name", "Age", "City"],
        ["John", "25", "Beijing"],
        ["Jane", "30", "Shanghai"]
    ],
    header_bold=True
)

✏️ Modify Table Cell

# Modify cell (1, 2) of the first table
update_table_cell(
    file_path="document.docx",
    table_index=0,
    row=1,
    col=2,
    new_text="New Content",
    font_name="Microsoft YaHei",
    font_size=12,
    is_bold=True
)

📁 Project Structure

mcp_word_edit/
├── 📄 server.py           # MCP server main file
├── 📋 requirements.txt    # Python dependencies
├── 🐳 Dockerfile          # Docker image configuration
├── 🐳 docker-compose.yml  # Docker Compose configuration
├── 🚫 .dockerignore       # Docker build ignore file
├── 📖 README.md           # Documentation
└── 📂 documents/          # Documents directory (Docker mount point)

🐳 Docker Information

💾 Data Persistence

When using Docker, mount a local directory to /documents in the container:

docker run -it --rm -v /path/to/your/docs:/documents mcp-word-commander:latest

📦 Image Details

⚡ Technical Features

🤝 Contributing

Issues and PRs are welcome!

Before submitting, please ensure:

• ✅ Code style consistency
• ✅ Necessary documentation included
• ✅ Related tests passed

📜 License

This project is licensed under the MIT License. See LICENSE for details.