### Install Dependencies and Serve PPTAgent Frontend

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Navigate to the UI directory, install Node.js dependencies, and start the development server for the PPTAgent frontend. The backend API endpoint is configured in `pptagent_ui/vue.config.js`.

```bash
cd pptagent_ui
npm install
npm run serve
```

--------------------------------

### Install and Run PPTAgent MCP Server

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Installs the PPTAgent using uv, sets necessary environment variables for model and API configuration, and starts the MCP server. This is used for generating presentations via an MCP server.

```bash
uv pip install pptagent
export PPTAGENT_MODEL=openai/gpt-4.1
export PPTAGENT_API_BASE=http://localhost:8000/v1
export PPTAGENT_API_KEY=your_key
uv run pptagent-mcp
```

--------------------------------

### Install PPTAgent Locally

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Installs the PPTAgent package with all optional dependencies for full functionality. Alternatively, it shows how to install the package in editable mode for local development.

```bash
pip install "pptagent[full]"
# or install locally
pip install -e ".[full]"
```

--------------------------------

### Start DeepPresenter Web UI

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Starts the DeepPresenter web interface application.

```bash
python webui.py
```

--------------------------------

### Initialize DeepPresenter CLI

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Runs the interactive setup wizard for the DeepPresenter CLI. This command helps in configuring the application for first-time use.

```bash
uvx pptagent onboard
```

--------------------------------

### Copy Configuration Files

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Copy example configuration files to be used by the application. These files are essential for setting up the application's behavior.

```bash
cp deeppresenter/config.yaml.example deeppresenter/config.yaml
cp deeppresenter/mcp.json.example deeppresenter/mcp.json
```

--------------------------------

### Install Python Dependencies and Playwright

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Installs the project in editable mode using uv, installs Playwright dependencies, and downloads the Chromium browser for Playwright. Also downloads a language identification model.

```bash
uv pip install -e .
playwright install-deps
playwright install chromium
modelscope download forceless/fasttext-language-id
```

--------------------------------

### Install uv Package Manager

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Installs the 'uv' package manager using a curl script. This is a prerequisite for managing Python dependencies.

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```

--------------------------------

### Start DeepPresenter Host Service

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Start the DeepPresenter host service in detached mode using Docker Compose. The web UI will be accessible at http://localhost:7861.

```bash
docker compose up -d
```

--------------------------------

### Run PPTAgent Backend

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Execute the backend server script for PPTAgent. This script is typically run after setting up the environment variables.

```python
python pptagent_ui/backend.py
```

--------------------------------

### Pull and Tag Docker Images (1ms.run Mirror)

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Pulls the DeepPresenter sandbox and host Docker images from the 1ms.run mirror and tags them for local use.

```bash
docker pull docker.1ms.run/forceless/deeppresenter-sandbox
docker pull docker.1ms.run/forceless/deeppresenter-host
docker tag docker.1ms.run/forceless/deeppresenter-sandbox deeppresenter-sandbox
docker tag docker.1ms.run/forceless/deeppresenter-host deeppresenter-host
```

--------------------------------

### Initialize Environment Variables for PPTAgent

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Set these environment variables before initializing the models for PPTAgent. Ensure you replace placeholders with your actual API keys and service endpoints.

```bash
export OPENAI_API_KEY="your_key"
export API_BASE="http://your_service_provider/v1"
export LANGUAGE_MODEL="openai/gpt-4.1"
export VISION_MODEL="openai/gpt-4.1"
export MINERU_API="http://localhost:8000/file_parse"
```

--------------------------------

### Generate Presentation with Title

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Generates a presentation from a single page with a specified title. The output is saved as a .pptx file.

```bash
uvx pptagent generate "Single Page with Title: Hello World" -o hello.pptx
```

--------------------------------

### Build Docker Images from Dockerfile

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Builds the DeepPresenter sandbox and host Docker images locally using their respective Dockerfiles.

```bash
docker build -t deeppresenter-sandbox -f deeppresenter/docker/SandBox.Dockerfile .
docker build -t deeppresenter-host -f deeppresenter/docker/Host.Dockerfile .
```

--------------------------------

### Generate Presentation with Attachments

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Generates a presentation including specified attachments like data files and PDFs, and a page range. The output is saved as a .pptx file.

```bash
uvx pptagent generate "Q4 Report" \
  -f data.xlsx \
  -f charts.pdf \
  -p "10-12" \
  -o report.pptx
```

--------------------------------

### Configure MCP Server in Claude/Cursor

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Configuration snippet for integrating PPTAgent as an MCP server within tools like Claude or Cursor. It specifies the command, arguments, and environment variables required to run the server.

```json
{
  "mcpServers": {
    "pptagent": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "pptagent",
        "pptagent-mcp"
      ],
      "env": {
        "PPTAGENT_MODEL": "openai/gpt-4.1",
        "PPTAGENT_API_BASE": "http://localhost:8000/v1",
        "PPTAGENT_API_KEY": "your_key"
      }
    }
  }
}
```

--------------------------------

### Cite DeepPresenter Environment (BibTeX)

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Use this BibTeX entry to cite the DeepPresenter work, which focuses on environment-grounded reflection for agentic presentation generation. This entry is for the arXiv preprint.

```bibtex
@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},
}
```

--------------------------------

### View Docker Container Logs

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Follows the logs of the running PPTAgent Docker container to monitor its activity and troubleshoot issues.

```bash
docker logs -f pptagent
```

--------------------------------

### Pull and Tag Docker Images (Docker Hub)

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Pulls the DeepPresenter sandbox and host Docker images from Docker Hub and tags them for local use.

```bash
docker pull forceless/deeppresenter-sandbox
docker pull forceless/deeppresenter-host
docker tag forceless/deeppresenter-sandbox deeppresenter-sandbox
docker tag forceless/deeppresenter-host deeppresenter-host
```

--------------------------------

### Pull and Run PPTAgent Docker Image

Source: https://github.com/icip-cas/pptagent/blob/main/pptagent/DOC.md

Pulls the latest PPTAgent Docker image and runs it in detached mode, mapping necessary ports and volumes. It configures environment variables for API keys and mounts the home directory for model caching.

```bash
docker pull forceless/pptagent:latest

# mapping home directory to /root to allow caching of models
docker run -dt --gpus all --ipc=host --name pptagent \
  -e OPENAI_API_KEY=$OPENAI_API_KEY \
  -e MINERU_API=$MINERU_API \
  -p 9297:9297 \
  -p 8088:8088 \
  -v $HOME:/root \
  forceless/pptagent

# set -e PULL=True to pull latest changes from the repository
# append /bin/fish to override the default command
```

--------------------------------

### Cite PPTAgent Project (BibTeX)

Source: https://github.com/icip-cas/pptagent/blob/main/README.md

Use this BibTeX entry to cite the PPTAgent paper published at EMNLP 2025. It includes details about authors, publication venue, and abstract.

```bibtex
@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."
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.