Slidev-AI is a web app that leverages LLM (Large Language Model) technology to make creating Slidev-based online presentations elegant and effortless. It is designed to help engineers and academics quickly produce content-focused, minimalist PPTs that are easily shareable online.
🎉 SlidevAI received the Best Application Award in ModelScope MCP&Agent Competition.
slidev-ai is a downstream implementation within the OpenMCP ecosystem, demonstrating how developers can build specialized agents using OpenMCP's powerful framework. This project serves as:
- A reference implementation for OpenMCP agent development
- A production-ready presentation generation solution
- A template for creating domain-specific AI agents
Check out the full demo on Bilibili:
For developers looking to create similar AI-powered applications, here's a comprehensive prompt you can use with LLM to generate a similar website project:
Please refer to Quick Start. For containerized deployment, see Deploy with Docker | 中文文档.
Two containers are recommended:
- backend: NestJS API on port 3001 (uses SQLite and Puppeteer)
- frontend: Vite+Vue static site served by Nginx on port 8080
We provide backend/Dockerfile, frontend/Dockerfile, and docker-compose.yml.
- Prepare environment variables (backend validates these at startup):
On Windows cmd (temporarily for current shell):
set OPENAI_API_KEY=your_key
set OPENAI_BASE_URL=https://api.openai.com/v1
set OPENAI_MODEL=gpt-4o-mini
Or create a .env at repo root:
OPENAI_API_KEY=your_key
OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_MODEL=gpt-4o-mini
- Build and start:
docker compose up -d --build
- Access:
- Frontend: http://localhost:8080
- Backend API: http://localhost:3001/api
Volumes: uploads (backend_data), presentations (presentation_data), and sqlite (sqlite_data) persist across restarts.
Notes:
- Frontend API endpoint is configured at build time via
VITE_DOMAIN,VITE_PORT, andVITE_ENABLE_HTTPS. The compose file defaults tolocalhost:3001. - Puppeteer dependencies are preinstalled in the backend image and run with
--no-sandboxflags. - If ports 8080 or 3001 are occupied, edit the
portssection indocker-compose.yml.
We welcome contributions from the community! Please see our Contribution Guidelines and Code of Conduct for details.
Slidev AI is open-source software licensed under the MIT License with additional terms for commercial use. See LICENSE for full details.
If you seek for tech support and deeper understanding of Slidev AI, please join our OpenMCP qq group:
"From ideas to presentations to content distribution — our mission is to shorten the distance." - The Slidev AI Team