PPTAgent
An Agentic Framework for Reflective PowerPoint Generation
What is PPTAgent?
PPTAgent is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to agentic framework for reflective powerpoint generation
An Agentic Framework for Reflective PowerPoint Generation
This server falls under the Business Applications category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- An Agentic Framework for Reflective PowerPoint Generation
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx pptagentConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use PPTAgent
PPTAgent (also known as DeepPresenter) is an agentic AI framework that autonomously generates PowerPoint presentations by analyzing reference materials, planning slide structure, designing layouts, and producing PPTX files with optional text-to-image asset creation. It supports document inputs such as Excel spreadsheets, PDFs, and Markdown files, and integrates web search via Tavily and PDF parsing via MinerU for research-backed slides. Users invoke it through a CLI, a web UI, or as an MCP server, making it suitable for quickly turning reports, data, or prompts into polished presentation decks.
Prerequisites
- Python 3.10+ with uv package manager or pip
- Node.js and npm (for html2pptx rendering component)
- An LLM API key configured in deeppresenter/config.yaml (OpenAI, Anthropic, or compatible)
- Optional: TAVILY_API_KEY for web search enrichment
- Optional: MINERU_API_KEY for high-quality PDF parsing
Install PPTAgent via uvx (quickest method)
The uvx method downloads and runs PPTAgent without a permanent install. Run the onboard command first to configure your LLM provider and other settings interactively.
uvx pptagent onboardGenerate your first presentation
Use the generate command with a text prompt and optional input files. The -o flag sets the output PPTX file path.
uvx pptagent generate "Q4 Sales Report" -f data.xlsx -f charts.pdf -o report.pptx
# Simple one-liner:
uvx pptagent generate "Single Page with Title: Hello World" -o hello.pptxInstall from source for advanced use
For Docker deployment, MCP server use, or development, clone the repository and install all components including the html2pptx renderer.
git clone https://github.com/icip-cas/PPTAgent.git
cd PPTAgent
uv pip install -e .
playwright install-deps
npm install --prefix deeppresenter/html2pptxConfigure the LLM and optional services
Copy the example config file and edit it with your API keys and preferred model. The config.yaml controls LLM provider, text-to-image model, and offline mode.
cp deeppresenter/config.yaml.example deeppresenter/config.yaml
# Edit deeppresenter/config.yaml:
# Set llm provider, api_key, and optionally t2i_model
# Add to environment if using Tavily or MinerU:
export TAVILY_API_KEY=tvly-...
export MINERU_API_KEY=...Launch the web UI (optional)
For a browser-based interface, use Docker Compose to start the full stack. The web UI is accessible at http://localhost:7861.
docker pull forceless/deeppresenter-sandbox
docker pull forceless/deeppresenter-host
docker compose up -d
# Then open http://localhost:7861Configure as an MCP server
PPTAgent supports MCP integration through its mcp.json configuration. Copy the example and add the server to your MCP client config.
cp deeppresenter/mcp.json.example deeppresenter/mcp.json
# MCP client config:
{
"mcpServers": {
"pptagent": {
"command": "uvx",
"args": ["pptagent", "serve"],
"env": {
"TAVILY_API_KEY": "tvly-..."
}
}
}
}PPTAgent Examples
Client configuration
MCP client configuration for Claude Desktop using PPTAgent's serve command via uvx.
{
"mcpServers": {
"pptagent": {
"command": "uvx",
"args": ["pptagent", "serve"],
"env": {
"TAVILY_API_KEY": "tvly-...",
"MINERU_API_KEY": "your-mineru-key"
}
}
}
}Prompts to try
Example prompts for generating presentations via an MCP client connected to PPTAgent.
- "Generate a 10-slide presentation about transformer architectures with diagrams"
- "Create a Q4 business review deck from the attached spreadsheet data"
- "Make a slide deck comparing three cloud providers: cost, reliability, and features"
- "Build a company onboarding presentation with a welcome slide, team overview, and key tools"Troubleshooting PPTAgent
uvx pptagent onboard fails with 'No module named playwright'
Run 'playwright install-deps' after installing PPTAgent from source. The uvx distribution may require installing playwright separately with 'pip install playwright && playwright install'.
Generated PPTX file has missing images or blank slides
Check that the html2pptx npm package was installed correctly with 'npm install --prefix deeppresenter/html2pptx'. For offline mode, set 'offline_mode: true' in config.yaml to skip network-dependent asset generation.
PDF input parsing produces poor-quality text extraction
Set MINERU_API_KEY in your environment to use the MinerU PDF parsing service, which produces significantly better results than the local fallback parser for complex PDFs with tables and figures.
Frequently Asked Questions about PPTAgent
What is PPTAgent?
PPTAgent is a Model Context Protocol (MCP) server that agentic framework for reflective powerpoint generation It connects AI assistants to external tools and data sources through a standardized interface.
How do I install PPTAgent?
Follow the installation instructions on the PPTAgent GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with PPTAgent?
PPTAgent works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is PPTAgent free to use?
Yes, PPTAgent is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
PPTAgent Alternatives — Similar Business Applications Servers
Looking for alternatives to PPTAgent? Here are other popular business applications servers you can use with Claude, Cursor, and VS Code.
n8n
★ 189.1kA comprehensive MCP server that provides full control over n8n automation workflows through natural language. It offers 43 tools for managing workflows, executions, credentials, and data tables, with safety features like write-mode protection and dou
LobeHub
★ 77.5k🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
Jeecgboot
★ 46.4kAI 低代码平台，「低代码 + 零代码」双模式驱动：低代码一键生成前后端代码，零代码 5 分钟搭建系统，AI Skills 一句话画流程、设计表单、生成整套系统。内置 AI聊天、知识库、流程编排、MCP插件等，兼容主流大模型。引领「AI 生成 → 在线配置 → 代码生成 → 手工合并->AI修改」开发模式，消除 Java 项目 80% 的重复工作，提效而不失灵活。
CowAgent
★ 44.7kCowAgent (chatgpt-on-wechat) 是基于大模型的超级AI助理，能主动思考和任务规划、访问操作系统和外部资源、创造和执行Skills、通过长期记忆和知识库不断成长，比OpenClaw更轻量和便捷。同时支持微信、飞书、钉钉、企微、QQ、公众号、网页等接入，可选择DeepSeek/OpenAI/Claude/Gemini/ MiniMax/Qwen/GLM/LinkAI，能处理文本、语音、图片和文件，可快速搭建个人AI助理和企业数字员工。
Minds Platform
★ 39.2kPlatform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.
Astrbot
★ 32.8kAI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨
Browse More Business Applications MCP Servers
Explore all business applications servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up PPTAgent in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use PPTAgent?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.