word-agent-mcp
Enables AI agents to edit Word documents using track changes, allowing lawyers to review each modification via Word's revision interface.
README
word-agent-mcp
让 AI agent 以**修订模式（Track Changes）**编辑 Word 文档的 MCP server —— 律师版 "Claude Code for Word" 的文档操作层。
核心理念
程序员审代码用 diff，律师审文书用修订。本项目让 agent 的每一处修改都以标准 Word 修订（红线删除 + 下划线插入）写入 .docx，律师在 Word「审阅」界面逐条接受/拒绝——agent 永远不能未经审阅直接改动文书。
修订直接写入 OOXML（w:ins/w:del），不需要安装 Word、跨平台（Mac 开发、Windows 交付零移植）。
架构
src/word_agent_mcp/
├── server.py # MCP 工具层（read_document / replace_text / insert_paragraph / ...）
├── revisions.py # OOXML 修订标记核心（w:ins / w:del 生成、run 拆分、修订扫描）
└── backend/
├── base.py # WordBackend 抽象接口（按 Windows COM 能力超集设计）
└── ooxml.py # 文件后端：python-docx + lxml，跨平台
# （规划）com.py：Windows COM 后端，实时操控打开中的 Word
工具层只依赖 WordBackend 接口。阶段 2 在 Windows 上补一个 COM 后端即可获得"实时看着 agent 改文档"的体验，工具层零改动。
快速开始
uv sync # 安装依赖（自动准备 Python 3.11+）
uv run pytest # 跑测试
uv run python examples/demo.py # 生成演示合同并修订，用 Word 打开看红线
接入 Claude Code
claude mcp add word-agent -- uv --directory /path/to/word-agent-mcp run word-agent-mcp
或在项目 .mcp.json 里：
{
"mcpServers": {
"word-agent": {
"command": "uv",
"args": ["--directory", "/path/to/word-agent-mcp", "run", "word-agent-mcp"]
}
}
}
然后直接对话：「读一下 ~/Desktop/租赁合同.docx，把租金改成 9000 并加一条逾期解约条款」。
MCP 工具
| 工具 | 说明 |
|---|---|
read_document |
读取全部段落（index / style / 有效文本） |
replace_text |
修订式替换；不唯一时要求更长上下文或 replace_all |
insert_paragraph |
在指定段后修订式插入新段落 |
delete_paragraph |
修订式删除段落 |
list_revisions |
列出文档全部修订（自检 / 汇报用） |
安全设计：每次编辑前自动留 .bak 备份；所有修改必须过 Word 审阅这一关。
已知限制（MVP）
- 只处理正文段落，表格/页眉页脚/脚注暂不支持
- 匹配区间内含制表符、图片等复杂 run 结构时会明确报错（不会静默破坏文档）
- 批注（comments）待加入
- Windows 上 Word 正在打开的文件会被锁定：请让 agent 改副本，或关闭文档后编辑
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.