Skip to content

afikbae/paper2video / .claude/skills/paper-to-video

paper-to-video

Verified

End-to-end workflow that turns a research paper (PDF or LaTeX) into a narrated video — paper -> slides -> voice-over -> assembled video, optionally with other videos spliced in. Use when the user wants a supplementary video, paper video, talk video, video abstract, or "make a video from my paper", or asks to rebuild/update such a video in this repo.

Paper to video

All scripts live in scripts/ at the repo root. Run them with uv run from the repo root (the venv is managed by uv; run uv sync and uv run playwright install chromium once if .venv is missing). Every script takes the PROJECT folder as its first argument and has --help.

Stages

#StageCommandProduces
0Create projectuv run scripts/init_project.py projects/NAME --paper PAPER.pdffolder layout, paper/ extraction
1Read paperread paper/paper.md (or paper.tex), look at paper/pages/*.pngyour understanding
2Scriptwrite narration.jsonvoice-over text per segment
3Slideswrite slides/NN-name.html, then uv run scripts/render_slides.py P --checkout/slides/*.png
4Voiceuv run scripts/generate_voice.py Paudio/*.wav
5Timelineuv run scripts/make_timeline.py P, then edit timeline.jsonordered items
6Videouv run scripts/assemble_video.py P --draft --sheet, review, then without --draftout/video.mp4

Stage details live in the sibling skills: paper-to-slides (stages 1-3), voiceover (stage 4), video-editing (stages 5-6 and anything involving other videos).

How to run it

  1. Ask only what you cannot infer. Before writing, settle: target length (default 3 min of narration for a supplementary video), venue limits (size in MB, minutes, anonymity), audience, and whether other videos (demo, screen recording, user study footage) should be included and where. If the user already said, don't ask.
  2. Script before slides. Draft narration.json first (one segment per slide, ids NN-short-name), ~150-170 spoken words per minute. Show the user the script and outline; it is cheap to change now and expensive after TTS.
  3. Slides follow the script. One idea per slide; the narration carries detail, the slide carries the headline, one figure or 3-4 short bullets. Same id as the narration segment.
  4. Render and look. Always run render_slides.py --check and open the PNGs to look at them yourself. Fix every reported issue.
  5. Voice. Use --provider say or --provider silent for a free timing draft if the user has no API key yet; real TTS needs OPENROUTER_API_KEY in .env. Report clips flagged LOW.
  6. Assemble a draft with --draft --sheet, read the contact sheet PNG to check the order and content, then build the final. Report length, size, and any max_mb / max_minutes warnings.

Rules

  • Never invent results, numbers or claims; everything on slides and in narration must come from the paper.
  • Anonymous submissions: no author names, affiliations, logos, or identifying footage (faces, lab names, window titles). Ask before including footage that might identify people.
  • Keep generated media out of git (out/, audio/*.wav are ignored); the source of truth is narration.json, slides/*.html, timeline.json and media/.
  • Re-running is cheap: voice clips and video segments are cached and only rebuilt when their inputs change.
canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill.Stars: 198kUpdated Claude CodeCodexVerifiedInstall