Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/alohays/paper2pr/write-speaker-notesnpx skills add alohays/paper2pr --skill write-speaker-notesgit clone --depth 1 https://github.com/alohays/paper2prWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/alohays/paper2pr/write-speaker-notes)<a href="https://agentmods.dev/skills/alohays/paper2pr/write-speaker-notes"><img src="https://agentmods.dev/badge/skills/alohays/paper2pr/write-speaker-notes.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00070 | $0.03849 |
| Opus 5 | $0.00035 | $0.01924 |
| Sonnet 5 | $0.00014 | $0.00770 |
| Haiku 4.5 | $0.00007 | $0.00385 |
Grade A, and why
write-speaker-notes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 362 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Speaker Notes (Presentation Script) Workflow
Generate a verbatim reading script for a deck. The presenter reads these notes as-is during the talk, so every note is the actual sentences to say aloud, in the language the deck declares.
CRITICAL: These are presentation scripts (대본), NOT talking points. Full spoken sentences that flow naturally when read aloud.
The deck's genre decides what a good script is. Never guess the genre: read the resolved profile first (Phase 0B) and apply the matching genre section below. The mechanics (where notes live, the three privacy layers) are the same for every genre.
Phase 0: Pre-Flight Checks
0A. Identify Target File
Decks live at Quarto/<genre>/<name>.qmd. Resolve the argument rather than
building a path by hand:
python3 scripts/deckpath.py [Deck] --field qmd
A bare name works; genre/name disambiguates if two genres ever share one.
If no argument, ask the user. python3 scripts/deckpath.py --list shows what
exists.
0B. Read the Deck's Premises (profile + deck.yml)
Everything the script depends on is declared, not guessed. Read both, first thing:
python3 scripts/deckprofile.py [Deck] # resolved config, JSON
cat "$(python3 scripts/deckpath.py [Deck] --field config)" # the raw deck.yml
From the resolved JSON:
profile- which genre section below applies:paper-review,lecture, orinvited-talknotes_language-ko(280 Hangul syllables/min) oren(130 words/min)speaking_min- the minutes the presenter actually talksprior_session- for a lecture in a series: the previous session's title, week, date and presenter. The opening note names it (see Genre: Lecture)sources- what technical claims in the notes are checked against
From the raw deck.yml: audience.assumes, audience.size, audience.prior
and delivery - review-agent context the resolved JSON does not carry; the
invited-talk section depends on it.
Budget = speaking_min x 280 Hangul syllables (ko) or x 130 words (en).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 362 lines · 70 tokens per session scan A c564610ed9d3
write-speaker-notes is a skill published in the GitHub repository alohays/paper2pr (5 stars, last pushed 4d ago), licensed MIT. It adds 70 tokens to every session and 3,849 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
beamer-academic
Generate high-quality academic Beamer slides from a thesis/paper (PDF, Word, or LaTeX source). Compatible with Claude Code and Codex. Supports thesis defense, proposal presentations, and conference talks. Built-in layout library with 13 professional page types, 5 color schemes, and interactive editing loop. Use when…
econ-slides
Build professional academic economics slide decks in LaTeX/Beamer: turn a paper into a conference, seminar, or job-market talk; build discussant slides; polish or shorten an existing deck. Use when the user mentions slides, presentation, talk, deck, beamer, seminar, conference, job talk, job market presentation…
course-generate
Generate a complete course — learning objectives, syllabus, slides (LaTeX), scripts, and assessments — using the ADDIE multi-agent pipeline. Use when the user asks to "create a course on X", "generate teaching materials for X", "build an undergraduate/graduate course on X", or similar end-to-end course-authoring…
slide-evaluate
Evaluate generated course slides from two perspectives — Program Chair (academic rigor) and Test Student (clarity, engagement). Use when the user wants feedback on existing slide materials, asks "review these slides", "evaluate this course", or after running course-generate and wanting quality feedback.
article-writing
Write articles, guides, blog posts, tutorials, newsletter issues, and other long-form content in a distinctive voice derived from supplied examples or brand guidance. Use when the user wants polished written content longer than a paragraph, especially when voice consistency, structure, and credibility matter.
eli5
Explain research, papers, or technical ideas in plain English with minimal jargon, concrete analogies, and clear takeaways. Use when the user says "ELI5 this", asks for a simple explanation of a paper or research result, wants jargon removed, or asks what something technically dense actually means.