Slidev AI is an AI-powered web application designed to streamline the creation of Slidev-based online presentations from text. It targets engineers and academics, enabling them to rapidly transform dense, well-structured textual content into minimalist, content-focused presentations for efficient online distribution, thereby shortening the gap between ideas and their dissemination.

How It Works

This project leverages Large Language Model (LLM) technology to automate the generation of Slidev presentations. The core approach involves converting high-density textual material into visually organized slides, ideal for scenarios like team meeting reports or internal technical presentations. As a downstream implementation within the OpenMCP ecosystem, it demonstrates building specialized agents using a powerful framework.

Quick Start & Requirements

• Primary Install/Run: Containerized deployment using Docker is recommended.
• Prerequisites: Requires environment variables for OpenAI API key, base URL, and model (e.g., OPENAI_API_KEY, OPENAI_BASE_URL, OPENAI_MODEL).
• Dependencies: Backend utilizes NestJS, SQLite, and Puppeteer; Frontend uses Vite+Vue and Nginx.
• Access: Frontend accessible at http://localhost:8080, Backend API at http://localhost:3001/api.
• Resources: Dockerfiles and docker-compose.yml are provided. See Deploy with Docker documentation for details. A demo video is available here.

Highlighted Details

• Awarded "Best Application" in the ModelScope MCP&Agent Competition.
• Serves as a reference implementation for OpenMCP agent development.
• Acts as a production-ready presentation generation solution.
• Provides a template for creating domain-specific AI agents.

Maintenance & Community

Contributions are welcomed via the provided Contribution Guidelines and Code of Conduct. For technical support and deeper understanding, users can join the OpenMCP Developer Group (QQ).

Licensing & Compatibility

The project is licensed under the MIT License with additional terms specified for commercial use. Full details are available in the LICENSE file. Compatibility for commercial use hinges on these additional terms.

Limitations & Caveats

Slidev AI is designed for converting well-structured text into slides and cannot interpret or restructure content that the user does not fully comprehend. It is not suitable for tasks requiring deep contextual analysis or adaptation, such as interpreting academic papers or generating complex company annual reports.