local_fire_departmentHoneystax
search⌘K
loginLog Inperson_addSign Up
layers
HONEYSTAX TERMINAL v1.0
HomeNewsSavedSubmit
Back to the live board
P

PPTAgent

MCP Server

An Agentic Framework for Reflective PowerPoint Generation

Copy the install, test the workflow, then decide if it earns a permanent slot.

3,752
Why nowMoving now

Fresh repo activity plus visible builder pull. This is the kind of tool people test before it turns obvious.

DecisionHigh-conviction move

Copy the install, test the workflow, then decide if it earns a permanent slot.

Trial costMedium lift

Testable in one sitting, but you will likely touch real infra or local setup before you know if it sticks.

Risk35/100

GitHub health 50/100. no security policy. 7 open issues make this testable, but not something to trust blind.

What You Are Adopting

AI Agent

Universal

Model

Multiple

Build Time

Minutes

Test This In Your Stack

One command inClean rollbackLow commitment
settingsRegistryAdds a named entry to Claude config. One command to remove.

Fastest way to find out if PPTAgent belongs in your setup.

Copy the install command, run a real test, and back it out cleanly if it slows you down.

Try now
claude mcp add pptagent -- npx pptagent

Run this first. You will know quickly if the workflow earns a permanent slot.

Back out
claude mcp remove pptagent

No messy cleanup loop. If it misses, remove it and keep moving.

Install Location

~/  └─ .claude.json    └─ mcp_servers/      └─ pptagent ← registers here

About

An Agentic Framework for Reflective PowerPoint Generation. An open-source mcp server for the AI coding ecosystem.

README

🌐 Language
English | 简体中文 | 繁體中文 | 日本語 | 한국어 | हिन्दी | ไทย | Français | Deutsch | Español | Italiano | Русский | Português | Nederlands | Polski | العربية | فارسی | Türkçe | Tiếng Việt | Bahasa Indonesia | অসমীয়া
test.mp4

Contact 📫

The main contributor of this repo is a Master's student graduating in 2026, feel free to reach out for collaboration or opportunities.

本仓库的主要贡献者是一名 2026 届硕士毕业生,欢迎联系合作或交流机会。

News 📅

  • [2026/01]: We support freeform and template generation support PPTX export, offline mode now! Context management is added to avoid context overflow.
  • [2025/12]: 🔥 Released V2 with major improvements - Deep Research Integration, Free-Form Visual Design, Autonomous Asset Creation, Text-to-Image Generation, and Agent Environment with sandbox & 20+ tools.
  • [2025/09]: 🛠️ MCP server support added - see MCP Server for configuration details
  • [2025/09]: 🚀 Released v2 with major improvements - see release notes for details
  • [2025/08]: 🎉 Paper accepted to EMNLP 2025!
  • [2025/05]: ✨ Released v1 with core functionality and 🌟 breakthrough: reached 1,000 stars on GitHub! - see release notes for details
  • [2025/01]: 🔓 Open-sourced the codebase, with experimental code archived at experiment release

Usage 📖

Important

  1. All these API keys, configurations, and services are required.
  2. Agent Backbone Recommendation: Use Claude for the Research Agent and Gemini for the Design Agent. GLM-4.7 is also a good choice in open-source models.
  3. Offline mode is supported with limited capabilities (see Offline Setup below).

1. Environment Configuration

  • Create configuration files (from project root):

    cp deeppresenter/deeppresenter/config.yaml.example deeppresenter/deeppresenter/config.yaml
    cp deeppresenter/deeppresenter/mcp.json.example deeppresenter/deeppresenter/mcp.json
  • Online setup:

    • MinerU: Apply for an API key at mineru.net. Note that each key is valid for 14 days.
    • Tavily: Apply for an API key at tavily.com.
    • LLM: Set your model endpoint, API keys, and related parameters in config.yaml.
  • Offline setup:

    • MinerU: Deploy the MinerU server by following the instructions at MinerU docker guide
    • Config switch: Set offline_mode: true in config.yaml to avoid loading network-dependent tools (e.g., fetch, search).
    • MinerU endpoint: Set MINERU_API_URL in mcp.json to your local MinerU service URL

2. Service Startup

Build docker images: docker compose build

  • From Docker Compose:

    docker compose up -d
  • Running locally:

    cd deeppresenter
    pip install -e .
    playwright install-deps
    playwright install chromium
    npm install
    npx playwright install chromium
    python webui.py

Tip

🚀 All configurable variables can be found in constants.py.

Case Study 💡

  • Prompt: Please present the given document to me.

图片1 图片2 图片3 图片4 图片5 图片6 图片7 图片8 图片9 图片10
  • Prompt: 请介绍小米 SU7 的外观和价格

图片1 图片2 图片3 图片4 图片5 图片6
  • Prompt: 请制作一份高中课堂展示课件,主题为“解码立法过程:理解其对国际关系的影响”

图片1 图片2 图片3 图片4 图片5 图片6 图片7 图片8 图片9 图片10 图片11 图片12 图片13 图片14 图片15

Contributors 🌟

Force1ess/
Force1ess
Puelloc/
Puelloc
Dnoob/
Dnoob
Sadahlu/
Sadahlu
hongyan/
hongyan
KurisuMakiseSame/
KurisuMakiseSame
Eliot
Eliot White
EvolvedGhost/
EvolvedGhost
ISCAS-zwl/
ISCAS-zwl
James
James Brown
JunZhang/
JunZhang
Open
Open AI Tx
Sense_wang/
Sense_wang
SuYao/
SuYao
Zhenyu/
Zhenyu

Star History Chart

Citation 🙏

If you find this project helpful, please use the following to cite it:

@inproceedings{zheng-etal-2025-pptagent,
    title = "{PPTA}gent: Generating and Evaluating Presentations Beyond Text-to-Slides",
    author = "Zheng, Hao  and
      Guan, Xinyan  and
      Kong, Hao  and
      Zhang, Wenkai  and
      Zheng, Jia  and
      Zhou, Weixiang  and
      Lin, Hongyu  and
      Lu, Yaojie  and
      Han, Xianpei  and
      Sun, Le",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.emnlp-main.728/",
    doi = "10.18653/v1/2025.emnlp-main.728",
    pages = "14413--14429",
    ISBN = "979-8-89176-332-6",
    abstract = "Automatically generating presentations from documents is a challenging task that requires accommodating content quality, visual appeal, and structural coherence. Existing methods primarily focus on improving and evaluating the content quality in isolation, overlooking visual appeal and structural coherence, which limits their practical applicability. To address these limitations, we propose PPTAgent, which comprehensively improves presentation generation through a two-stage, edit-based approach inspired by human workflows. PPTAgent first analyzes reference presentations to extract slide-level functional types and content schemas, then drafts an outline and iteratively generates editing actions based on selected reference slides to create new slides. To comprehensively evaluate the quality of generated presentations, we further introduce PPTEval, an evaluation framework that assesses presentations across three dimensions: Content, Design, and Coherence. Results demonstrate that PPTAgent significantly outperforms existing automatic presentation generation methods across all three dimensions."
}

@misc{zheng2026deeppresenterenvironmentgroundedreflectionagentic,
      title={DeepPresenter: Environment-Grounded Reflection for Agentic Presentation Generation},
      author={Hao Zheng and Guozhao Mo and Xinru Yan and Qianhao Yuan and Wenkai Zhang and Xuanang Chen and Yaojie Lu and Hongyu Lin and Xianpei Han and Le Sun},
      year={2026},
      eprint={2602.22839},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2602.22839},
}

Tech Stack

GoPythonClaudeLLMDocker
Open Live ProjectAudit Repo

Reviews0

Log in to write a review.

ActiveLast commit today
bug_report7open issues
Submitted January 4, 2025

auto_awesomeYour strongest next moves after PPTAgent