remarkable
/remarkable
Renderuje dokument Markdown lub Org do PDF do czytania na reMarkable Paper Pure w klasycznym stylu prac informatycznych, przez pandoc i tectonic.
- Autor
- wunki ↗
- Repozytorium
- dotfiles ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 20.10.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 3
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 20.10.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Renderuje dokument Markdown lub Org do PDF do czytania na reMarkable Paper Pure w klasycznym stylu prac informatycznych, przez pandoc i tectonic.
Wyzwalacze z opisu
Oryginalny opis (EN)
Render a Markdown or Org document to a PDF for reading on Petar's reMarkable Paper Pure, in the clean classic CS/Lisp-paper aesthetic (Computer Modern Book, small-caps numbered sections, monochrome code, links as footnotes), using pandoc and tectonic. Use when the user asks for a PDF for the reMarkable, a "paper PDF", "Lisp paper aesthetic", a clean PDF of a Markdown or Org document, or to regenerate one after editing the source. Don't use for slides, HTML output, or when the user wants a specific other template.
Jak działa
- 1. Check the toolchain
- 2. Survey non-ASCII characters
- 3. Make the print copy
- 4. Decide the numbering
- 5. Title, author, date
- 6. Render
- 7. Verify and hand off
Podgląd SKILL.md
---
name: remarkable
description: Render a Markdown or Org document to a PDF for reading on Petar's reMarkable Paper Pure, in the clean classic CS/Lisp-paper aesthetic (Computer Modern Book, small-caps numbered sections, monochrome code, links as footnotes), using pandoc and tectonic. Use when the user asks for a PDF for the reMarkable, a "paper PDF", "Lisp paper aesthetic", a clean PDF of a Markdown or Org document, or to regenerate one after editing the source. Don't use for slides, HTML output, or when the user wants a specific other template.
---
Turn a Markdown or Org document into a PDF that looks like a classic CS
paper and reads well on e-ink: the MIT AI memo look, sized for the
reMarkable. New Computer Modern Book, 10pt article, small-caps section
headings, monochrome code, URLs as footnotes. Pandoc reads both formats;
the examples below say `DOC.md`, and an Org source only changes the two
spots called out in steps 4 and 5.
The source document is never modified. All transformation happens in a print
copy under the session scratchpad.
## 1. Check the toolchain
```bash
which pandoc tectonic
```
Both required. If either is missing, ask the user to install it
(`brew install pandoc tectonic`); do not install it for them.
## 2. Survey non-ASCII characters
Latin Modern Mono lacks box-drawing and arrow glyphs; they render as blanks.
Find what the document uses:
```bash
LC_ALL=C grep -o '[^ -~]' DOC.md | LC_ALL=C sort | LC_ALL=C uniq -c | LC_ALL=C sort -rn
```
`LC_ALL=C` on every stage is load-bearing: macOS locale collation silently
merges distinct Unicode punctuation, and the survey lies without it.
Curly quotes, en/em dashes, and ellipses are fine (Latin Modern has them).
Arrows and box-drawing characters must be transliterated in the print copy.
## 3. Make the print copy
Write to the scratchpad, never next to the source:
```bash
sed -E -e 's/──>/-->/g' -e 's/└─>/\\->/g' -e 's/→/->/g' \
-e 's/─/-/g' -e 's/└/\\/g' -e 's/│/|/g' -e 's/├/+/g' \
DOC.md > "$SCRATCH/DOC-print.md"
```
Extend the substitution list for whatever the survey found. Order matters:
replace multi-character sequences (`──>`, `└─>`) before their parts.
## 4. Decide the numbering
Check whether headings already carry their own numbers (`## 3. The publish
API`). Double numbering looks broken ("14.1 1. Publish one normal post").
- Headings are unnumbered prose: pass `--number-sections`.
- Headings self-number at every level: drop `--number-sections`.
- Mixed (numbered subsections under prose sections): keep `--number-sections`
and strip the manual numbers in the print copy, e.g. add
`-e 's/^### [0-9]+\. /### /'` to the sed above (for Org sources the
heading marker is stars: `-e 's/^\*\* [0-9]+\. /** /'`). Only do this when
the generated numbers still line up with any cross-references to the old
ones.
## 5. Title, author, date
- Markdown opening with a single `#` title: pass
`--shift-heading-level-by=-1`, so the H1 becomes the paper title and `##`
become sections. Org with `#+title:`: the title is already metadata and
`*` headings are already sections, so drop the shift flag entirely.
Anything else: pass `-M title="..."`.
- No author. Petar is the only reader; a byline adds nothing.
- Date: today, spelled out ("August 6, 2026").
## 6. Render
The header is `assets/remarkable-header.tex`, next to this file. Use it via
the skill's base directory; do not inline a new one.
The page is sized to the screen of Petar's reMarkable Paper Pure (10.3in,
1872x1404, 226 PPI), so text renders at true point size with no zooming:
6.21in x 8.28in, 0.6in side margins (room for pen annotations), 0.55in top,
0.7in bottom with `footskip=0.3in`. The explicit footskip is what keeps the
page number off the bezel: LaTeX's default (~0.42in) is as deep as a thin
bottom margin, which parks the folio on the screen edge. The header sets
New Computer Modern in its Book weight: heavier strokes made for
low-contrast e-ink, where Latin Modern's hairlines go faint. Verbatim is
`\small` (~9pt against the 10pt body) so code stays prominent on e-ink;
that fits ~78-character lines in the column. The same PDF reads fine on a
laptop and prints acceptably; no separate profile.
```bash
pandoc "$SCRATCH/DOC-print.md" -o DOC.pdf \
--pdf-engine=tectonic \
--shift-heading-level-by=-1 \
--number-sections \
--syntax-highlighting=monochrome \
-H "$SKILL_DIR/assets/remarkable-header.tex" \
-V documentclass=article \
-V fontsize=10pt \
-V "geometry:paperwidth=6.21in,paperheight=8.28in,hmargin=0.6in,tmargin=0.55in,bmargin=0.7in,footskip=0.3in" \
-V links-as-notes=true \
-M date="..."
```
Notes:
- The command above is for Markdown; for an Org source drop
`--shift-heading-level-by=-1` (step 5) and adjust any number-stripping
sed to star headings (step 4). Everything else is identical.
- `--syntax-highlighting` needs pandoc >= 3.10; older pandoc calls it
`--highlight-style`.
- The header loads `fancyvrb` explicitly. Without it, documents that have no
fenced code blocks fail on `\fvset` with "Undefined control sequence".
- Tectonic downloads packages on first run; allow a generous timeout.
- The remarkable header loads New Computer Modern by OTF filename through
fontspec. Tectonic's bundled `newcomputermodern.sty` is too old for the
`[book]` package option; `\setmainfont{NewCM10-Book.otf}` works because
tectonic resolves font files from its bundle by name.
- For handbook-style documents where each section is a chapter meant to be
read alone (checklists with lessons, curricula), add
`-H "$SKILL_DIR/assets/sectionbreak.tex"` as a second header: every
section then starts on a fresh page. Leave it off for flowing documents
like design plans.
## 7. Verify and hand off
- Overfull/underfull hbox warnings under ~2pt are cosmetic; ignore them.
Larger overfulls usually mean a code line wider than ~78 characters; wrap
it in the print copy.
- `ls -la` the PDF, then `open` it for the user.
- Be honest in the handoff: without poppler installed you verified the build,
not the typography; the user is the visual check.
- Offer the exact regeneration command (or add it to a Makefile if the
project has one and the user wants repeatability).
Pliki w paczce (3)
- SKILL.md 6 KB
- assets/remarkable-header.tex 1 KB
- assets/sectionbreak.tex 1 KB
Wersje
W bazie nie ma innych skilli o nazwie remarkable.
Migawka bazy z 13.11.2026 · wersja paczki 31f0312087a7 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100send-to-remarkable
Tworzy PDF w rozmiarze reMarkable z artykułu lub tekstu i wysyła go przez SSH teraz albo w kolejce. - 86md-to-pdf-closed-source
Buduje PDF z Markdowna za pomocą pandoc, zamieniając tekst na plik gotowy do wysyłki. - 100pandoc
Konwertuje dokumenty między formatami Markdown, DOCX, HTML, PDF, RST, EPUB i LaTeX za pomocą wiersza poleceń pandoc. - 100md-to-doc
Zamienia pliki Markdown na dokumenty Word (.docx) lub PDF z tabelami, blokami kodu i spisem treści, przez pandoc i zapasowe silniki PDF. - 100pdf-generation
Składa rozdziały w markdownie w książkę lub dokumentację w PDF, przy użyciu pandoc i weasyprint. - 100remarkable-setup
Konfiguruje krok po kroku tablet reMarkable i Maca do logowania kluczem SSH przez USB i Wi-Fi, oraz sprawdza lub naprawia połączenie.