# Turn research papers into Chinese Nature-style PPT decks

Researchers waste hours manually building slide decks from papers for journal clubs and seminars. This skill automates the process, producing Chinese PowerPoint decks with figure-led evidence slides in Nature-style narrative arc.

## Install

```bash
npx skillstore add community contribution, refactored into static/dynamic layers/yuan1z0825-nature-paper2ppt
```

## Metadata

- - Slug: yuan1z0825-nature-paper2ppt
- - Version: 2.0.0
- - Author: Community contribution, refactored into static/dynamic layers
- - GitHub username: Yuan1z0825
- - License: MIT
- - Repository: https://github.com/Yuan1z0825/nature-skills/tree/main/skills/nature-paper2ppt
- - Ref: main
- - Supported tools: Claude, Codex, Claude Code
- - Risk level: low
- - Risk factors: filesystem, external\_commands
- - Quality score: 78
- - Quality tier: bronze
- - Public page: https://skillstore.pages.dev/skills/yuan1z0825-nature-paper2ppt
- - Manifest: https://skillstore.pages.dev/api/skills/yuan1z0825-nature-paper2ppt/manifest

## Capabilities

- Converts scientific papers into 10-16 slide Chinese PPTX presentations
- Classifies paper type \(discovery, methods, resource, clinical, materials, review\) and selects appropriate narrative arc
- Extracts key figures and tables as evidence, not decoration
- Writes Chinese slide titles, bullets, captions, takeaways, and speaker notes
- Generates real .pptx files using python-pptx with 16:9 layout
- Runs self-review and corrective revision loop with severity-graded defect list
- Produces a QA report and asset manifest alongside the deck

## Use Cases

- Journal club presentation: PhD student prepares a 15-minute journal club presentation from a Nature paper and needs a Chinese deck with key figures and speaker notes.
- Lab group meeting slides: Research lab member summarizes a discovery paper for weekly group meeting with evidence-led slides.
- Thesis defense preparation: PhD candidate builds a comprehensive academic deck from multiple referenced papers for thesis defense.

## Prompt Templates

### Basic paper-to-PPT request

```
Convert this Nature paper into a Chinese PPTX presentation for journal club: [paste PDF path or text]. Identify the paper type and create 12 slides with figure-led evidence.
```

### Specify paper type and audience

```
This is a methods paper about a new AI tool. Create a 14-slide Chinese presentation for a 20-minute lab meeting. Emphasize problem-to-solution narrative and include the main architecture figure.
```

### Review paper with evidence map

```
Convert this review paper into a Chinese PPTX using evidence-map structure. Group findings by theme. Include comparison tables where the paper provides explicit data. Generate speaker notes for each slide.
```

### Multi-paper seminar deck

```
I need a 20-slide Chinese academic deck combining these three papers on [topic]. Use a thematic structure with evidence from each paper. Include a synthesis slide at the end.
```

## Limitations

- Default language is simplified Chinese; other languages require user request
- Rendered slide previews require LibreOffice or similar headless renderer
- Full supplementary material processing is opt-in, not automatic
- Skill provides instructions only; execution depends on the AI assistant's tool availability

## Best Practices

- State the paper type detection to the user before designing slides so they can correct it early
- Use the Terminology Ledger to keep technical terms consistent across all slides and speaker notes
- Select only figures that carry the argument rather than decorating with every available figure

## Anti Patterns

- Do not fabricate numbers, mechanisms, or figure details not present in the source paper
- Do not force every paper into the same slide template; adapt the structure to the paper type
- Do not deliver slides where text is expected to overflow or be clipped; split or shorten instead

## Security Audit

- - Safe to publish: true
- - Audited at: 2026-06-24T06:03:08.981\+00:00
- - Summary: This skill is a documentation-only skill consisting of markdown instruction files for an AI assistant to convert scientific papers into PowerPoint presentations. No executable code, scripts, or binaries are present. The static analyzer's 222 findings are all false positives: 'Ruby/shell backtick execution' matches are markdown triple-backtick code fence delimiters \(\`\`\`\), 'Weak cryptographic algorithm' matches are triggered by documentation text patterns, 'Path traversal sequence' matches are legitimate relative path references \(e.g., ../\_shared/\), 'System reconnaissance' matches are the word 'system' in legitimate documentation context, and the 'Python archive libraries' match is a legitimate zipfile reference for PPTX package validation. No prompt injection attempts were detected. The skill is safe to publish.

## Stats

- - Views: 0
- - Downloads: 19
- - Favorites: 0
- - Popularity score: 0