Aut_Sci_PPt skill
A skill that turns a scientific paper PDF or a numbered outline into a finished .pptx with sharp formulas and figures.
- 123 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The Python package reads PDFs and writes `.pptx` files to disk, and free-form text is sent to an external model API for parsing.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/aut-sci-pptDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The Python package reads PDFs and writes `.pptx` files to disk, and free-form text is sent to an external model API for parsing.
This is third-party code. Review the repository files before installing.
What it does
Aut_Sci_PPt is a skill and Python package that builds a `.pptx` deck for a defense, seminar or research proposal. The input is a paper PDF or text with `1. Title` sections and `- point` bullets. The PDF is parsed with PyMuPDF, figures are extracted at high resolution, and LaTeX formulas are rendered into transparent PNGs. The layout engine keeps text from overflowing the slide or overlapping figures. The theme is built in: a deep blue primary color and a red accent for the main novelty of the work. Free-form text can optionally be turned into structure through an Anthropic or OpenAI model when an API key is set in the environment.
Who it is for. Master and PhD students who need a defense, seminar or research proposal deck built from a finished paper PDF.
Good fit when
- The paper PDF exists and seminar slides have to be built in one evening.
- An admissions talk about you and your achievements is already drafted as sections and bullet lists.
- The deck has many LaTeX formulas that must stay sharp on a projector.
- Figures from the paper are needed on slides at high resolution, without blurry raster crops.
Not a fit when
- The design must follow a university or company template exactly, and the project has one built-in theme.
- The slides are not about a scientific paper: the logic is built for the academic format.
- Text cannot be sent to an external AI service, and free-form text goes to a model API for parsing.
Example request
Build a defense deck from this PDF, keep the formulas sharp and highlight the main novelty of the work in red.Limitations
Python 3.10 or newer is required. The `pip install python-pptx pyyaml pymupdf Pillow` command from the README installs only dependencies, and the package itself is used from a copy of the repository. The skill follows 13 layout rules from `SKILL.md`: it forbids cutting figures out of the PDF as low-quality raster images and warns when an operation will take more than 10 seconds. The theme sets Microsoft YaHei, KaiTi and Times New Roman fonts, and without them PowerPoint substitutes others. The last commit in the repository dates to July 2026.
How to disable. Remove the skill folder from the agent skills directory and uninstall the Python dependencies if you added them separately. API keys sit in environment variables, so clear them yourself.
FAQ
How should the input text be formatted?
Number sections as `1. Title` and start bullets with `- `. The parser does not recognize Markdown headings with `##`.
How does PDF mode differ from text mode?
PDF mode runs through `PaperWorkflow` and pulls text and figures from the paper with PyMuPDF. Text mode takes a ready outline through `create_ppt`.
Is an AI key required?
Only for parsing free-form text: `ai_parser.py` reads `ANTHROPIC_API_KEY` or `OPENAI_API_KEY`. Marked-up text goes through a separate parser with no model call.
What file comes out?
A finished `.pptx` in 16:9 (13.33 by 7.5 inches). The skill does not hand back intermediate Markdown.
How is the main novelty of the paper highlighted?
The skill looks for it and marks it in red `#EE0000`, while the main theme color is `#1E3A5F`.
Discussion
No questions or reviews yet.
Related
SkillsA skill that turns an article, report or notes into a deck of full-slide images and assembles them into a pptx
Codex and Claude Code skill that turns a research paper into a literature report deck with real figures and an evidence check
A Chinese-language skill that turns PDF, DOCX, web pages and Markdown into PPTX presentations with real editable shapes.
A skill that builds decks with gpt-image-2: each slide is drawn as a full layout and packed into .pptx, optionally mimicking a reference template
A skill that builds a deck from a topic or article, or restyles an old one from a reference image, keeping the text and editable slides
A skill for talks: outlines a deck around the audience's train of thought, picks images and video, and checks finished slides before you present
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| Aut_Sci_PPt (this card) | no | 123 | |
| Thesis Defense PPTX | no | 280 | keeps the design of your PowerPoint template; Aut_Sci_PPt uses a built-in theme and renders formulas itself |
| Academic Presentations Skill | no | 1,512 | builds slides as the argument of a talk; Aut_Sci_PPt adds a Python layout engine with PDF and figure parsing |
README in short
The README is brief: a feature list, dependency install and a Python example. Details sit in `SKILL.md`, `references/rules.md` and `USER_GUIDE.md`, and a finished sample deck is in the `example/` folder.
README badge
Original SKILL.md
---
name: aut-sci-ppt
description: Use when converting academic paper PDFs or structured text into professional, high-standard PowerPoint (PPTX) presentations for defense or seminars.
---
# Aut_Sci_PPt (Academic Auto-PPT Agent)
A specialized tool for generating professional academic presentations directly from paper content or structured outlines.
## Core Behavioral Rules (The 13 Laws)
1. **Format**: Use `1. Title` for chapters and `- Point` for bullets.
2. **Markdown**: 🚫 DO NOT use `##` for slide titles; it is not recognized by the parser.
3. **Innovation**: Identify the core innovation of the paper and highlight it in **Red**.
4. **Imagery**: Use HD extraction (600 DPI) for figures; minimum width >= 300px.
5. **No Scrapping**: 🚫 PROHIBITED to scrap low-quality bitmaps from PDF streams.
6. **Formulas**: Render LaTeX formulas as high-quality transparent PNGs.
7. **Transparency**: All generated formula/media assets must have transparent backgrounds.
8. **Feedback**: Inform the user if an operation (like PDF parsing) will take >10 seconds.
9. **Final Output**: 🚫 DO NOT output intermediate Markdown; generate and provide the `.pptx` directly.
10. **Colors**: Use `#1E3A5F` (Primary Blue) and `#EE0000` (Highlight Red).
11. **Layout**: Ensure zero text-overflow or figure-text overlap.
12. **Professionalism**: Keep communication brief and technical; skip AI pleasantries.
## Usage
### Simple Text Input
```python
from aut_sci_ppt import create_ppt
create_ppt("""
主题：[Title]
申请人：[Name]
1. [Section Title]
- [Content]
""", "output.pptx")
```
### PDF to PPT (Academic Workflow)
```python
from aut_sci_ppt import PaperWorkflow
workflow = PaperWorkflow()
workflow.process_pdf("paper.pdf", output_pptx="seminar.pptx")
```