paper2html — Claude Code skill: what it does, how to install
paper2html converts academic papers from arXiv, LaTeX or PDF into self-contained Chinese reading pages with KaTeX math and verified evidence.
no license ★ 475 updated 2026-10-13
npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a claude-codeWhat paper2html does
This workflow converts academic papers into standalone HTML documents designed for thorough reading without reopening original files. It handles arXiv identifiers, OpenReview entries, local TeX packages, and standard PDF documents. The process relies on dedicated helper scripts rather than pure prompt generation.
Execution starts with bootstrap_paper2html.py creating the workspace and documentation notes. If TeX sources exist, tex_inventory.py catalogs equations and notations. Meanwhile, optimize_paper_images.py rescales extracted figures and packages them directly into data URI strings, ensuring the finished page functions independently of external image hosting.
The resulting page follows a defined narrative structure: summary, method flow, experimental evidence, bounded takeaways, and review viewpoints. Local paths remain excluded from output pages to prevent accidental private data leaks when publishing.
What it helps with
- turn arXiv papers into self-contained HTML reading sheets
- bundle TeX formulas and figures into single-file documents
- extract experimental reproducibility details from LaTeX sources
- generate structured review analyses across multiple evaluation roles
Example prompts for the agent
- Process paper 2504.07952 from arXiv into an index.html sheet using paper2html
- Analyze unpacked TeX sources and package all figures with data URIs
- Validate the generated research HTML file before shipping it
How to install paper2html
1Skills CLI (Vercel)
npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a claude-code2Manually: put the files in place
~/.claude/skills/paper2html/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/paper2html/
1Skills CLI (Vercel)
npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a codex2Manually: put the files in place
.agents/skills/paper2html/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a cursor2Manually: put the files in place
.cursor/skills/paper2html/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install cnfjlhj/ai-collab-playbook skills/full/paper2html2Manually: put the files in place
.github/skills/paper2html/SKILL.md1Skills CLI (Vercel)
npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a gemini-cli2Platform CLI
gemini skills install https://github.com/cnfjlhj/ai-collab-playbook --path skills/full/paper2htmlManually: put the files in place
.devin/skills/paper2html/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about paper2html
How to install paper2html in Claude Code?
Run npx skills add cnfjlhj/ai-collab-playbook --skill paper2html -a claude-code in your terminal.
What does paper2html do?
It orchestrates scripts to translate academic papers and TeX sources into self-contained HTML deep-reading pages with KaTeX formatting.
Does paper2html require local TeX sources?
No, it accepts arXiv IDs or PDFs directly, though it prioritizes TeX sources for better formula and outline extraction when they are present.