How do I use ppt-word-gen?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@ppt-word-gen Generate a Word simulation report based on the built-in template" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

de en es ja ko ru zh

ppt-word-gen

by muli4561

Overview Schema Related Servers Score Discussions

Python

Hybrid

ppt-word-mcp

AI-driven editable PPTX and standardized Word simulation test report generation service
A single lightweight service provides MCP (Model Context Protocol) Server, REST API, and Web Demo on the same port.

🚀 MCP (Model Context Protocol) Service

ppt-word-gen natively implements the standard MCP protocol (supporting both Streamable HTTP and STDIO transport channels), empowering various Agent clients (such as Claude Code, Claude Desktop, Cursor, Cline, OpenWebUI, etc.) to autonomously complete document format extraction, user confirmation, asynchronous queued generation, semantic multi-round revision, and secure file download.

1. Client Integration Configuration (`.mcp.json`)

The project root already includes .mcp.json. Agent tools can automatically discover and load the service when opening this directory.

Method A: HTTP Transport Mode (recommended, suitable for container deployment / LAN services / local services)

Create or edit .mcp.json in the project root:

{
  "mcpServers": {
    "ppt-word-gen": {
      "type": "http",
      "url": "http://127.0.0.1:8000/mcp"
    }
  }
}

If the service has Bearer Token authentication enabled (configured with PPT_WORD_GEN_TOKEN), you can add request headers:

{
  "mcpServers": {
    "ppt-word-gen": {
      "type": "http",
      "url": "http://YOUR_SERVER_IP:8000/mcp",
      "headers": {
        "Authorization": "Bearer ${PPT_WORD_GEN_TOKEN}"
      }
    }
  }
}

# 项目级注册（写入当前目录 .mcp.json）
claude mcp add --transport http --scope project ppt-word-gen http://127.0.0.1:8000/mcp

# 用户全局注册
claude mcp add --transport http --scope user ppt-word-gen http://127.0.0.1:8000/mcp

Method B: STDIO Transport Mode (invoke directly on the local machine without starting an HTTP port)

In the client configuration, start directly via a Python process:

{
  "mcpServers": {
    "ppt-word-gen": {
      "command": "python",
      "args": ["-m", "ppt_word_gen.mcp_stdio"]
    }
  }
}

2. MCP Core Collaborative Workflow

[用户输入 / 来源资料 / 模板]
             │
             ▼
[可选：upload_file / create_upload_ticket 暂存来源文件与 Word 模板]
             │
             ▼
[preview_word_report_format 提取模板默认格式（字体、字号、行距、多级标题、编号规范）]
             │
             ▼
[向用户展示预览并获取确认 ➔ 获得绑定格式哈希的 confirmation_token]
             │
             ▼
[generate_presentation / generate_word_report 发起异步生成任务]
             │
             ▼
[wait_generation_task / get_generation_task 轮询进度状态]
             │
             ▼
[get_artifact 获取 24 小时有效的安全签名 ResourceLink 下载链接]
             │
             ▼
[可选：revise_presentation / revise_word_report 进行多轮语义增量修订]

3. Complete MCP Tool Set (15 Tools)

Tool Name	Description	Key Parameters
`list_generation_profiles`	List document types that can be generated, supported formats, and binary upload conventions	None
`preview_word_report_format`	Extract the template default format (or built-in template) and apply modifications; return format information and `confirmation_token`	`template_upload_id`, `custom_format`
`generate_presentation`	Create an editable PPTX asynchronous generation task	`prompt`, `format`, `source_upload_ids`, `model`, etc.
`generate_word_report`	Submit a Word report task (must include `confirmation_token`)	`title`, `report_type`, `confirmation_token`, `source_upload_ids`
`get_generation_task`	Query the stage, percentage progress, errors, and artifact status of a PPT or Word async task	`task_id`, `task_type`
`wait_generation_task`	Wait up to 55 seconds; returns immediately when the task completes, or returns the latest status on timeout	`task_id`, `task_type`, `timeout_seconds`
`cancel_generation_task`	Request cancellation of an async generation task that has not yet finished	`task_id`, `task_type`
`get_artifact`	After task completion, obtain a 24-hour secure signed download direct link with a standard ResourceLink	`task_id`, `task_type`
`upload_file`	Base64 inline upload for small files (no more than 5MB)	`filename`, `content_base64`, `purpose`
`create_upload_ticket`	Create a one-time PUT upload address for large files (avoid large content occupying Agent context)	`filename`, `purpose`, `max_bytes`
`list_business_templates`	List built-in and enterprise-customized PPT/Word business specification templates	None
`register_business_template`	Register enterprise business content and style specification templates	`template_id`, `name`, `spec`
`delete_business_template`	Delete custom business templates (built-in templates are protected and cannot be deleted)	`template_id`
`revise_presentation`	Use an existing PPTX as context and generate a new version according to natural language instructions	`parent_task_id`, `revision_prompt`
`revise_word_report`	After user confirmation of the format, use an existing DOCX as the base to incrementally generate a new version	`parent_task_id`, `confirmation_token`, `revision_instructions`

4. MCP Rules and Asset Resources (4 Resources)

Agents can directly read specification guides and design constraints:

ppt-word://rules/workflow: Complete MCP interaction and format confirmation specification
ppt-word://rules/presentation: PPT Master design and layout specification
ppt-word://rules/word-report: Simulation and engineering Word report layout standards
ppt-word://templates/catalog: Business template catalog and specification quick reference

Related MCP server: docforge-mcp

📁 Project Structure

ppt-word-gen/
├── ppt_word_gen/              # Python 核心业务包
│   ├── app.py                 # FastAPI 入口（Demo、REST API、MCP Streamable HTTP）
│   ├── mcp_server.py          # MCP Server 实现（15 工具、4 资源、双向协议）
│   ├── mcp_stdio.py           # MCP STDIO 运行入口
│   ├── tasks.py               # PPT 异步任务队列与状态机
│   ├── report_tasks.py        # Word 异步任务队列与状态机
│   ├── report_agent.py        # Word 报告生成 Agent
│   ├── report_documents.py    # Word 文档排版、格式继承与 GBK 修复引擎
│   ├── word_format.py         # Word 格式解析、修改与防篡改 Token 校验
│   ├── signed_tokens.py       # HMAC-SHA256 签名下载与上传票据
│   ├── pptmaster.py           # PPT Master 脚本适配与执行
│   ├── upload_store.py        # 临时文件上传与生命周期管理
│   ├── task_store.py          # SQLite 持久化任务存储
│   └── config.py              # 全局配置读取
├── assets/word_templates/     # 内置 Word 模板（CID629 电驱系统联合仿真 v1.5）
├── skills/ai-simulation-report/# 仿真报告标准契约与校验规则
├── static/                    # 前端 Web Demo 页面
├── tests/                     # 41 项单元测试与集成测试
├── Dockerfile                 # 容器构建文件（支持联网/离线两种模式）
├── docker-compose.yml         # 生产/测试容器编排配置
├── build-overlay.ps1          # 企业级解密与离线 wheel 构建脚本
├── requirements.txt           # 依赖清单
├── .env.example               # 环境变量参考模板
└── .mcp.json                  # MCP 配置文件

💻 Local Quick Start

1. Install Dependencies and Run Tests

# 安装依赖
python -m pip install -r requirements.txt

# 运行全套 41 项自动化测试
python -m unittest discover -s tests -v

2. Start the Service

# 启动 HTTP + MCP 服务（默认监听 0.0.0.0:8000）
python -m ppt_word_gen

Web Demo page: http://127.0.0.1:8000/demo
OpenAPI interactive documentation: http://127.0.0.1:8000/docs
MCP HTTP endpoint: http://127.0.0.1:8000/mcp
Health check: http://127.0.0.1:8000/health

🐳 Docker / WSL Containerized Deployment

1. Standard Deployment

In an Ubuntu / WSL terminal:

cd /path/to/ppt-word-gen
cp .env.example .env.compose

# 按需修改 .env.compose（接入真实模型时设置 MOCK_LLM=0 并填写 API Key）
sudo service docker start
docker compose --env-file .env.compose up -d --build --wait

2. Enterprise Offline Wheel Build

In environments with enterprise data leak prevention (DLP) or intranet isolation, execute in Windows PowerShell:

pwsh ./build-overlay.ps1

This script will automatically copy files securely via file stream and invoke OFFLINE_INSTALL=1 to perform a local offline Wheel build.

3. Service Status Check and Stop

# 查看容器状态与健康指标
docker compose ps
curl http://127.0.0.1:8000/health

# 停止容器（数据保留在 ppt-word-gen-data 卷中）
docker compose stop

⚙️ Environment Variable Reference

Copy .env.example to .env or .env.compose:

Variable Name	Default Value	Description
`LLM_BASE_URL`	`https://dashscope.aliyuncs.com/compatible-mode/v1`	OpenAI-compatible API Base URL
`LLM_API_KEY`	None	Large language model API Key
`LLM_MODEL`	`qwen3.7-plus`	Default creation model name
`MOCK_LLM`	`1`	`1` enables full end-to-end self-test demo without an API Key; set to `0` to connect a real model
`PPT_WORD_GEN_TOKEN`	empty	Bearer Token authentication; leave empty for no authentication
`PUBLIC_BASE_URL`	`http://127.0.0.1:8000`	Base URL for artifact downloads and large file uploads (set to the server's intranet IP for LAN calls)
`DOWNLOAD_SIGNING_SECRET`	empty (auto-generated)	HMAC signing secret for generating 24-hour temporary download credentials
`MAX_CONCURRENT_TASKS`	`2`	Maximum concurrent PPT processing tasks
`MAX_CONCURRENT_REPORT_TASKS`	`1`	Maximum concurrent Word processing tasks
`MCP_INLINE_UPLOAD_MB`	`5`	MCP Base64 inline upload limit (MB)
`MAX_UPLOAD_MB`	`20`	REST / large file Ticket upload limit (MB)

🔌 Core REST API

Request Method	Path	Description
`GET`	`/health`	Service health check (includes SQLite storage and task queue status)
`GET`	`/demo`	Built-in Web demo page
`GET`	`/api/v1/word-format`	Get built-in Word template default format and confirmation Token
`POST`	`/api/v1/word-format`	Extract custom template format, apply modifications, and generate a confirmation Token
`POST`	`/api/v1/report-tasks`	Submit Word simulation/technical report generation task
`GET`	`/api/v1/report-tasks/{id}`	Query Word report task progress and details
`GET`	`/api/v1/report-tasks/{id}/result`	Download the generated DOCX report
`POST`	`/api/v1/tasks`	Submit PPT presentation generation task
`GET`	`/api/v1/tasks/{id}`	Query PPT task progress and details
`GET`	`/api/v1/tasks/{id}/result`	Download the generated PPTX file
`POST`	`/api/v1/uploads`	Temporarily store binary source files or templates
`PUT`	`/api/v1/upload-tickets/{token}`	One-time large file direct upload interface
`GET`	`/api/v1/artifacts/{token}`	24-hour HMAC signed secure download interface
`POST`	`/mcp`	MCP Streamable HTTP protocol endpoint

📄 Open Source License

This project follows the MIT License.

This server cannot be installed

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Maintenance

ActivityMaintained

ResponsivenessSyncing

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

superdocs
app.superdocs
AI document editing for agents: draft, edit, export .docx/PDF. 37 MCP tools; agent self-signup.
Workplace & Productivity Text Summarization
21
Frenchie A
io.github.Lab94
OCR, transcription, file extraction, and image generation for AI agents via MCP.
AI & Machine Learning Image & Video Processing Speech Processing
72
Blitz Wings
io.github.scheur
AI presentation and report generation: slides, diagrams, PPTX export, live preview MCP App.
Design Tools Workplace & Productivity Data Visualization
2
emplusx
com.emplusx
Finished, on-brand .pptx and .docx from a brief - quality-gated by an agentic consulting team.
Workplace & Productivity Marketing Autonomous Agents
1

Related MCP Servers

Office Document Processing MCP
Workplace & Productivity App Automation
iflow-mcp
A
license
Not graded
quality
D
maintenance
Enables creation and editing of Word, Excel, and PowerPoint documents within MCP clients through AI assistants.
Updated 7 months ago (2026-02-08 02:32 UTC)
MIT
docforge-mcp
Workplace & Productivity Developer Tools
filhocf
A
license
B
quality
D
maintenance
Enables complete Office document lifecycle management for AI agents, including creation, editing, conversion, and templating of DOCX, XLSX, PPTX, PDF, and EML files.
Updated a month ago (2026-07-21 19:04 UTC)
40
MIT
pharos-ai-doc-genie
Developer Tools Workplace & Productivity AI & Machine Learning
huimingchen081-beep
A
license
A
quality
C
maintenance
Enables AI agents to generate real Office documents (.pptx, .docx, .xlsx) and source code from natural language via MCP protocol.
Updated 3 months ago (2026-06-13 00:54 UTC)
4
MIT
Egile MCP Slide Deck Generator
AI & Machine Learning Workplace & Productivity Content Management Systems
jpoullet2000
A
license
Not graded
quality
D
maintenance
Enables creation of professional PowerPoint presentations with AI-generated content and images, supporting multiple LLMs and image services via MCP protocol.
Updated 8 months ago (2026-01-12 06:07 UTC)
MIT

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/muli4561/ppt-word-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server