中文 | English

SlideAgent - Enables Everyone to Easily Create Professional Presentations

SlideAgent is an open-source, AI-driven presentation generator. Provide a topic or documents, and it produces an outline, slide content, and a themed deck with online preview and multi-format export.

If this helps you, please consider starring the repo ✨

🚧 Project Status

This project is still under active development and gradually improving:

• ✅ Core features are basically implemented
• 🔄 API interfaces may be adjusted
• 📝 Documentation is being continuously updated

Welcome to try it out and provide feedback!

🎉 News

• [2026/02] Add The Function of Online Search
• [2026/01] "See-What-You-Get" PPT Export Function

✨ Highlights

• AI PPT Generation — Automatically generates outlines, content, and designs
• Knowledge Base — Upload documents and generate PPTs from your own materials
• Online Preview & Editing — Preview slides in the browser and edit text directly
• Online Sharing — Generate share links with expiration settings
• Task Queue — Batch processing with backend queue
• PPTX Export Service — Dedicated export_tool service converts HTML slides to editable PPTX

✅ Current Status

• Content Editing — Edit slide text directly in the preview
• Online Preview — Real-time PPT preview in the browser
• Export — PDF / HTML / PPTX export (PPTX styles may be lost; improving)
• Task Status — Global persistence of agent task status

🧭 Roadmap

• Conversational Editing — Keep refining slides via dialogue
• Database Search Tool — Call knowledge base tools directly
• Multi-version Management — Save, compare, and rollback versions

🚀 Quick Start

Prerequisites

• Docker and Docker Compose
• Git

Run with Docker Compose

1. Clone the repository

   git clone https://github.com/Mrguanglei/SlideAgent.git
   cd SlideAgent

2. Configuration (Optional) Copy .env.example to .env and modify as needed:

   cp .env.example .env

   You can configure database and LLM API settings in the .env file.

3. Build and start services

   docker-compose up --build -d

Supported Model Providers API (Tested)

Provider/Series	Model	Status	Notes
Zhipu AI	GLM-5	✅ Supported
	GLM-4 Series	✅ Supported
	GLM-4-Plus	✅ Supported
	GLM-4-Flash	✅ Supported
DeepSeek	DeepSeek-V3	✅ Supported
	DeepSeek-V3.2	✅ Supported
	DeepSeek-R1	✅ Supported
MiniMax	MiniMax-Text Series	✅ Supported
ByteDance (Doubao)	Doubao-1.8	✅ Supported
	Doubao-pro	✅ Supported
Kimi	kimi-k2-turbo-preview	✅ Supported
Alibaba Cloud	Qwen-Max	✅ Supported
	Qwen-Plus	✅ Supported
Intern (Shanghai AI Lab)	intern-s1-pro	✅ Supported

🚧 TODO (Not Tested)

Provider/Series	Model	Status	Notes
OpenAI	GPT-4o	🚧 Not Tested	Compatibility mode pending verification
	GPT-4o-mini	🚧 Not Tested
Anthropic	Claude 3.5 Sonnet	🚧 Not Tested
	Claude 3 Opus	🚧 Not Tested
Baidu	ERNIE 4.0	🚧 Not Tested
Moonshot AI	Kimi k1.5	🚧 Not Tested
01.AI	Yi-Large	🚧 Not Tested
...	....	🚧 Not Tested

🧩 PPTX Export Service (HTML -> PPTX)

• Service: export_tool (FastAPI) runs independently and is started by Docker Compose
• Pipeline: Backend /api/ppt/export -> export_tool /api/export_tool/pptx
• Tech: Playwright (Chromium) renders HTML, dom-to-pptx converts to PPTX, with font embedding and icon assets
• Details: See export_tool/README.md for API usage and deployment notes

⚙️ Environment Variables

Create a .env file in the project root directory to override default configurations.

Variable	Default Value	Description
POSTGRES_DB	pptagent	Database name
POSTGRES_USER	pptagent	Database username
POSTGRES_PASSWORD	pptagent	Database password
DATABASE_URL	postgresql+asyncpg://...	Database connection string
PPTAGENT_API_BASE_URL	https://open.bigmodel.cn/api/paas/v4/	PPT generation LLM API base URL
PPTAGENT_API_KEY	your_api_key	PPT generation LLM API key
PPTAGENT_MODEL	glm-4-flash	PPT generation LLM model
KNOWLEDGE_LLM_BASE_URL	PPTAGENT_API_BASE_URL	Knowledge base LLM API base URL
KNOWLEDGE_LLM_API_KEY	PPTAGENT_API_KEY	Knowledge base LLM API key
KNOWLEDGE_LLM_MODEL	glm-4-flash	Knowledge base LLM model
KNOWLEDGE_EMBEDDING_MODEL	embedding-3	Knowledge base embedding model
KNOWLEDGE_UPLOAD_DIR	/tmp/knowledge_uploads	Knowledge base upload directory

📸 Screenshots

Home	Chat Generation
Knowledge Base	Global Search
Online Editing	Multiple Downloads

🛠️ Project Structure

.env.example         # Environment variables example
docker-compose.yml   # Docker compose configuration
README.md            # Project documentation

backend/             # Python backend (FastAPI)
├── services/        # Core services (export, sharing, knowledge base)
├── routers/         # API routes
├── database/        # Database models and CRUD
├── api_server.py    # FastAPI server entry point
├── requirements.txt # Python dependencies
└── Dockerfile

frontend/            # React frontend (Vite)
├── src/
│   ├── pages/       # Page components (Home, Knowledge, ShareView)
│   ├── components/  # Reusable components (Sidebar, Modals, etc.)
│   ├── lib/         # API requests, utility functions
│   └── types/       # TypeScript type definitions
├── package.json     # Node.js dependencies
├── vite.config.ts   # Vite configuration
└── Dockerfile

export_tool/         # Export service (PDF/PNG/HTML/PPTX)
├── app/             # FastAPI app and services
├── dom-to-pptx/     # HTML -> PPTX core library
├── fonts/           # Font assets for embedding
└── Dockerfile

🤝 Contributing

All forms of contributions are welcome! If you have any ideas or suggestions, feel free to submit Pull Requests or create Issues.

🙏 Acknowledgments

• Intern-S1 Pro - The Shanghai Artificial Intelligence Laboratory provides computing power support.
• shadcn/ui - Frontend UI component library.
• FastAPI - High-performance Python web framework.
• React - JavaScript library for building user interfaces.
• dom-to-pptx - The static library for exporting pptx

🖊️ Citation

@misc{2026SlideAgent,
    title={SlideAgent: Enables Everyone to Easily Create Professional Presentations},
    author={SlideAgent Contributors},
    howpublished = {\url{https://github.com/Mrguanglei/SlideAgent}},
    year={2026}
}

📄 License

This project is licensed under CC BY-NC-SA 4.0 (Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International).

• Attribution — You must give appropriate credit
• NonCommercial — You may not use the material for commercial purposes
• ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license

If this helps you, please consider starring the repo ✨