Skip to content

About

Reusable Codex skill for turning a paper PDF, preprint page, or source workspace into a readable static visualization site.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

Paper Visualization

Turn a paper PDF, preprint page, or source workspace into a readable research page, with figures, citations, and optional English and Chinese routes.

▶ Explore the interactive example · Get started · User guide · 简体中文

Paper Visualization: a clearer path through research. An annotated manuscript connects to question, method, and evidence.

MIT licence

A reading page you can explore

English interactive reading page: select question, method, or evidence to explore the annotated paper and its sources.

Select a passage, follow the reading path, inspect the source notes, and copy the software citation. Switch between English and Chinese editions.

▶ Open the English example

This custom static example uses the skill's public documentation as its source. It demonstrates the software workflow and presents no experimental findings.

From source material to a research page

Start with Keep visible
A paper PDF, preprint page, or source workspace The research question, method, evidence, and citation
Public figures and verified metadata The paper's identity and a route back to its sources
English and Chinese content Separate language routes with shared public assets

Use a structured JSON config to begin. When the paper needs a different composition, adapt the generated HTML, CSS, and JavaScript directly. Keep claim boundaries intact and choose figures that help explain the work.

Get started

For a fresh Codex installation:

git clone https://github.com/DELONG-L/Paper-Visualization-Skill.git
cd Paper-Visualization-Skill
mkdir -p "${CODEX_HOME:-$HOME/.codex}/skills"
cp -R skills/paper-visualization "${CODEX_HOME:-$HOME/.codex}/skills/"

If the skill is already installed, preserve local changes before replacing it. Provide your paper or source directory and ask:

Use $paper-visualization to turn
this paper into a readable
research page. Use the supplied
figures, preserve claim boundaries,
and add Chinese if needed.
Prefer to generate the scaffold directly?

The scaffold helper requires Python 3:

python3 skills/paper-visualization/scripts/bootstrap_paper_visualization.py ./my-paper --bilingual

Fill the generated configs and add the actual paper assets. The user guide covers PDF input, metadata, packaging, and alternate directories.

What the skill creates

  • A static research page: HTML, CSS, JavaScript, local figures, and a paper PDF when supplied.
  • Optional bilingual routes: docs/ and docs/zh/, with a config for each language.
  • Page metadata: descriptions, canonical URLs, language alternates, and a generated sitemap.
  • A portable site package: release/site-package/, ready for a static host. Keep private drafts and authoring sources outside the public site directory.
Try the example locally

Download or clone the repository, then run this command from its root:

python3 -m http.server 8123 --bind 127.0.0.1

Open the English example or the Chinese example.

Skill instructions · Config schema · Content workflow · Example source · Visual assets and sources · MIT licence

About

Reusable Codex skill for turning a paper PDF, preprint page, or source workspace into a readable static visualization site.

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages