paper-slides
Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says \"做PPT\", \"做幻灯片\", \"make slides\", \"conference talk\", \"presentation slides\", \"生成slides\", \"写演讲稿\", or wants beamer slides for a conference talk.
npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill paper-slides --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Generates a Beamer slide deck from a compiled paper, including speaker notes and a full talk script. It targets Beamer/PDF output and an editable PPTX export, with configurable venue, talk type, duration, and output directory. The workflow is designed to run after a paper-writing step and produces a slide outline before drafting content. It creates an outline file and then sequentially drafts slide content, finally compiling LaTeX into slides and optionally exporting to PPTX. It also supports generating a separate speaker notes backup and a full talk script. The process includes state persistence across phases and a Codex MCP review step when available.
How it works
- Uses a defined workflow with phases from input validation to outline, drafting, LaTeX compilation, external review, speaker notes, PPTX export, and full talk script.
- Phase 0 validates prerequisites, locates the paper, backs up existing slides, creates output directories, detects CJK content to adjust the LaTeX engine, and determines slide count from TALK_TYPE and TALK_MINUTES.
- Phase 1 reads the paper sections to build a slide-by-slide outline, with per-slide title, bullets, figure reference, speaker notes, and time allocation, saved as slides/SLIDE_OUTLINE.md. It stops at a checkpoint before drafting.
- Phase 2 drafts content for each slide following strict presentation rules (one idea per slide, max 6 lines, max 8 words per line, progressive disclosure, bold key numbers, etc.), producing LaTeX-ready frames and speaker notes per slide.
- Phase 3 generates slides/main.tex using a Beamer template, handles figure linking, and applies venue color schemes. It also provides commands to symlink figures into slides/figures.
- Phase 4 compiles the slides with latexmk, with a retry loop for common LaTeX errors, and verifies slide count matches the outline.
- Phase 5 optionally sends the outline and key frames to Codex MCP for review and stores the review results.
- Phase 6 ensures every slide has a \note{} block with speaker text, timing, and transitions, plus a separate slides/speaker_notes.md backup.
- Phase 7 exports an editable PPTX via python-pptx, creating slides that mirror Beamer frames, embedding figures, and transferring speaker notes. It can skip if python-pptx is unavailable.
- Phase 8 generates a comprehensive Talk Script document (slides/TALK_SCRIPT.md) with a full manuscript including per-slide sections, timing, transitions, and a time budget summary.
- State persistence files (SLIDES_STATE.json) track phase progress across the workflow.
When to use it
Use when the user asks for slides or presentations related to a conference talk, particularly with phrases like 做PPT, 做幻灯片, make slides, conference talk, presentation slides, 生成slides, 写演讲稿, or requests beamer slides for a conference talk.
What it can touch
- Directories and files: paper/ (input), slides/ (output), slides/figures/ (figures), slides/SLIDES_STATE.json, slides/SLIDE_OUTLINE.md, slides/main.tex, slides/generate_pptx.py, slides/presentation.pptx, slides/TALK_SCRIPT.md, slides/speaker_notes.md, slides/SLIDES_REVIEW.md, slides/SLIDES_STATE.json.
- Tools: Bash, Read, Write, Edit, Grep, Glob, Agent, mcp__codex__codex, mcp__codex__codex-reply.
Caveats
- The workflow is gated by explicit checkpoints; AUTO_PROCEED is false by default, requiring user confirmation to advance phases.
- Output depends on paper structure; if expected sections or figures are missing, outlines and content may omit those elements.
- PPTX export requires python-pptx installed; otherwise the export is skipped with a note.
# Paper Slides: From Paper to Conference Talk Generate conference presentation slides from: **$ARGUMENTS** ## Context This skill runs **after** Workflow 3 (`/paper-writing`). It takes a compiled paper and generates a presentation slide deck for conference oral talks, spotlight presentations, or poster lightning talks. Unlike posters (single page, visual-first), slides tell a **temporal story**: ea
What does the paper-slides skill do?
Generate conference presentation slides (beamer LaTeX → PDF + editable PPTX) from a compiled paper, with speaker notes and full talk script. Use when user says \"做PPT\", \"做幻灯片\", \"make slides\", \"conference talk\", \"presentation slides\", \"生成slides\", \"写演讲稿\", or wants beamer slides for a conference talk.
How do I install it?
Run `npx skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill paper-slides --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From brycewang-stanford/Auto-Empirical-Research-Skills, a repository with 3,244 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.