← Search

Research skills: Paper slide deck generator
Design 🧪 Install ran in our test sandbox · Nov 11, 2026 📌 Pinned version: commit 2b6f5bc
Install test only, on the date shown. This does not guarantee the code is safe.
🔎 Install prompt analysis · 2026-10-13 : installs packages · uses API keys or tokens
⚙ Needs: a Google/Gemini API key for image generation…

Research skills: Paper slide deck generator

Stylized slide-deck images from papers and articles via AI image generation — 17 aesthetic styles

At a glance
What
Stylized slide-deck images from papers and articles via AI image generation — 17 aesthetic styles
Cost
Free
Needs
a Google/Gemini API key for image generation (generous free tier, billed beyond it); without a key, only the degraded path below works (outline, figure extraction, prompts handoff); Node.js for the helper scripts (one-time npm install in the scripts folder); PyMuPDF for bitmap-heavy figures; Pillow (pip install Pillow), needed alongside PyMuPDF: when Gemini returns image/webp images, each slide image is converted to PNG with Pillow, and image generation fails without it. Licence: the repo declares no license, check its terms before reuse. Also note, before anything is installed: generated slides are AI images that garble math and numbers, and a deck now stops at a proofread gate until every generated slide carries a disposition, as described below.
Install
Copy the installer prompt below into your Muse — your agent does the rest.

Version:

Quick answer

Paper slide deck generator is a free Muse skill that turns papers and articles into stylized slide-deck images through AI image generation. It offers seventeen aesthetic styles, so the same research can become very different visual decks. That makes it handy for presenting academic work clearly, with a chosen look rather than a generic template.

⌁

Install

Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 11, 2026📌 Pinned version: commit 2b6f5bc

Skill Harbor note, October 26, 2026 (API-only revision): upstream still offers a web route, but this prompt does not recommend it and does not describe it, for trust reasons. The API route below is the only route described here. Before you start: tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead. Prerequisites: a Google/Gemini API key for image generation (generous free tier, billed beyond it); without a key, only the degraded path below works (outline, figure extraction, prompts handoff); Node.js for the helper scripts (one-time npm install in the scripts folder); PyMuPDF for bitmap-heavy figures; Pillow (pip install Pillow), needed alongside PyMuPDF: when Gemini returns image/webp images, each slide image is converted to PNG with Pillow, and image generation fails without it. Licence: the repo declares no license, check its terms before reuse. Also note, before anything is installed: generated slides are AI images that garble math and numbers, and a deck now stops at a proofread gate until every generated slide carries a disposition, as described below. Install "Research skills: Paper slide deck generator" for me. It turns any content, an article, a topic, or a paper, into stylized, shareable slide-deck IMAGES (not editable decks) via AI image generation, with automatic figure detection and extraction for PDF papers, and 17 aesthetic styles. Source: https://raw.githubusercontent.com/luwill/research-skills/2b6f5bc19d44faa59e06f99418af69abfe58200a/paper-slide-deck/SKILL.md Pinned version: install the code at commit 2b6f5bc19d44faa59e06f99418af69abfe58200a exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. Updated October 26, 2026: this prompt was regenerated after an upstream change. The headline change is a new proofread gate, Step 6.5, and it changes what shipping a deck means. The skill already warned that generated slides are AI images that garble math and numbers, and the previous prompt repeated that warning, but nothing enforced it. Now every slide recorded as generated (generated.json, written by generate-slides.py or by proofread.ts --record) must carry a proofread disposition before the merge scripts accept the deck. Both merge scripts refuse a deck, exit 1 and name the slides, when a recorded slide has no disposition or was regenerated after its disposition. Dispositions are ok (checked, nothing wrong) or check (still suspect after retries, listed as [CHECK] by the merge and in the Step 8 summary). Extracted-figure slides are never in generated.json and are not gated. The --allow-unproofread flag on the merge scripts overrides the refusal with a warning, only when you explicitly accept an unchecked deck, and the file says to fix the disposition rather than reaching for that flag. A second change is scope, not enforcement: the skill now says it transforms any content, an article, a topic, or a paper, into stylized, shareable slide images, not only academic papers and content, with figure detection and extraction kept for PDF papers. The proofread rule itself is the reason to regenerate: a deck that used to merge unchecked now stops at the gate, and an unchecked deck ships only on your explicit acceptance. 1. Download the SKILL.md file into a temporary folder. 2. Compare the file with the fingerprint from Skill Harbor's review of 2026-10-26: - SKILL.md: SHA-256 4f4f52f3ac372de73a4cf2e981d01c0a29ae07805916e4efee11b2d0a50960a0 If it does not match, the file changed since the review: do not install it yet, tell me what you found and let me decide. 3. Read the file and tell me in plain words: what the skill does when used, whether it connects to a service, whether it reads or sends any personal information or keys, and anything that seems unusual. Confirm these points from the file: - Generated slides are AI images, not editable decks: text, math and data are baked into pixels and will garble spelling, math symbols and numbers. That is why the proofread gate exists, and why this skill is not for a faithful academic talk where equations, numbers, tables and citations must stay exact, editable and projector-ready (for that, use scholar-slides instead). - The proofread gate, Step 6.5: for every generated slide, especially with equations, tables, key numbers or non-Latin text, visually check the PNG for spelling and wording, math and symbols (assume the model got them wrong until you confirm otherwise), and numbers and units matching the source exactly. Record every disposition with proofread.ts --mark (ok, or check with a note, for example a garbled sigma in an equation). Both merge scripts refuse a deck with a recorded slide that has no disposition or was regenerated after its disposition, exit 1 and name the slides. Fix the disposition rather than reaching for --allow-unproofread, which overrides the refusal with a warning only when I explicitly accept an unchecked deck. If garbling is found, regenerate that slide with a corrected or simplified prompt (spell risky terms phonetically, reduce text density, move exact numbers to a caption), at most 2 retries; if it still fails, flag it [CHECK] in the Step 8 summary and recommend replacing it with an extracted figure or table, simplifying it, or switching to scholar-slides. Never silently deliver a slide with garbled math or data. - The API path is the only path described here: it is recommended, with a GOOGLE_API_KEY or GEMINI_API_KEY, model gemini-3-pro-image, and a degraded no-key path (outline only, plus figure extraction and prompts handoff) if no key is available, without aborting. - Figure extraction limits: detect-figures.ts is heuristic, verify pages, especially the first page, and cross-check Figure 1 explicitly; extract-figure.ts renders the whole page to PNG, it does not auto-detect or crop a single figure, so on a two-column page you get both columns, and isolation needs --crop or a manual visual crop before apply-template.ts. PyMuPDF is the reliable fallback for bitmap-heavy pages. - Costs and missing licence: Gemini API use is billed beyond the generous free tier; the repo declares no license, so check its terms before reuse; the scripts need a one-time npm install, plus PyMuPDF and Pillow as required, not optional. 4. When I say yes, copy SKILL.md and its helper files into my skills folder, in a folder named "luwill-research-skills-paper-slide-deck": SKILL.md, the references/ folder (including the 17 styles), the scripts/ folder (generate-slides.py, merge-to-pptx.ts, merge-to-pdf.ts, detect-figures.ts, extract-figure.ts, apply-template.ts, proofread.ts, package.json, package-lock.json), and the learnings/ and tests/ folders as shipped. 5. Check the copy opens correctly (valid frontmatter, files in place) and tell me what was installed and where. In your report, please note the missing license, the Gemini API cost question, and the new proofread gate: no generated slide ships without a disposition unless I explicitly accept an unchecked deck. GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Rules: only touch the temporary folder and the install target. If something seems off, stop and ask me.

🔒 Security, provenance and what this touches
🧭 What this touches
  • It can download and install software packages.
  • It works with an API key or a token that you provide.
Install prompt analysis · 2026-10-13
🔒 Security and provenance
📌 Pinned version: commit 2b6f5bc · 2026-10-13
🧪 Install ran in our test sandbox · Nov 11, 2026
👁 Observed during the install test · Nov 11, 2026 : no network contact
🔎 Install prompt analysis · 2026-10-13 : installs packages · uses API keys or tokens
What the scan found — excerpts from the install prompt
  • installs packages — Prerequisites: a Google/Gemini API key (generous free tier — billed beyond it); Node.js for the helper scripts · Prérequis : une clé API Google/Gemini (niveau gratuit généreux — facturé au-delà) ; Node.js pour les scripts d
  • uses API keys or tokens — Prerequisites: a Google/Gemini API key (generous free tier — billed beyond it); Node.js for the helper scripts · Prérequis : une clé API Google/Gemini (niveau gratuit généreux — facturé au-delà) ; Node.js pour les scripts d
🔎 Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Wanna be extra careful?

The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?

How to check a skill before installing →

✓

Verification certificate

Dated facts about what was checked for this listing, and how. This is not a safety certification.

Pinned version
2b6f5bc · 2026-10-13
Sandbox install test
Install ran in our test sandbox · Nov 11, 2026
Creator identity
Not verified
Fingerprint watch
Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
Install prompt analysis
2026-10-13 : installs packages · uses API keys or tokens

Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify

?

Questions

How do I install a skill?

Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.

Where does my money go?

Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.

What does the ✓ next to a creator’s name mean?

It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.