PPTAgent

Name: PPTAgent
Author: icip-cas

@icip-cas

代码仓库

关于 PPTAgent

Please run `uv pip install pptagent` first

#pptagent #presentation-generator #evaluation-framework

基本信息

分类

其他

传输方式

stdio

发布者

icip-cas

提交者

less force

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "pptagent": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "pptagent",
        "--with",
        "python-pptx@git+https://github.com/Force1ess/python-pptx@219513d7d81a61961fc541578c1857d08b43aa2a",
        "pptagent-mcp"
      ],
      "env": {
        "PPTAGENT_MODEL": "openai/gpt-4.1",
        "PPTAGENT_API_BASE": "http://localhost:8000/v1",
        "PPTAGENT_API_KEY": "your_key"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is PPTAgent?

PPTAgent is an MCP server that automatically generates presentations from documents. It uses a two-step process inspired by human presentation creation, and includes PPTEval, a comprehensive evaluation framework for assessing presentation quality across multiple dimensions.

How to use PPTAgent?

Get started quickly with the pre-built Docker image or configure the MCP server as described in the documentation. Refer to DOC.md for detailed Docker and MCP server instructions.

Key features of PPTAgent

Dynamic content generation with seamlessly integrated text and images
Smart reference learning without manual annotation
Comprehensive quality assessment via PPTEval across Content, Design, and Coherence
Two-phase approach: Analysis and Generation
Pre-built Docker image for easy setup

Use cases of PPTAgent

Automatically generate a presentation from a document (e.g., product pages, research reports)
Create visually cohesive slides for business or academic talks
Evaluate the quality of generated presentations using PPTEval

FAQ from PPTAgent

Is PPTAgent open source?

Yes, PPTAgent is open source. The codebase is available on GitHub and the model and data are released on HuggingFace.

How does PPTAgent generate presentations?

PPTAgent uses a two-phase process: an Analysis phase that learns patterns from reference presentations, and a Generation phase that produces structured outlines and visually cohesive slides.

What is PPTEval?

PPTEval is a comprehensive evaluation framework that assesses presentations across three dimensions: Content (accuracy and relevance), Design (visual appeal and consistency), and Coherence (logical flow).

How can I quickly start using PPTAgent?

The quickest way is to use the pre-built Docker image. See the Docker instructions in the documentation for details.

Is there a research paper about PPTAgent?

Yes, a paper titled "PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides" has been accepted to EMNLP 2025 and is available on arXiv.