菜单
Download a document from a reMarkable device/cloud via remarquee (rmapi-backed) and render it to an annotated PDF via `remarquee rmdoc render-v6`/`render-legacy`, optionally extracting the last N pages into a smaller PDF. Use when asked to export/render/download a reMarkable notebook/journal/document to PDF (full doc or page subset).
Research a technology topic deeply using web search, Kagi Assistant, defuddle, and PDF collection, then write a comprehensive textbook-style research report in a docmgr ticket workspace and copy the full report to the Obsidian vault. Use when a user asks to research a topic, write a research report, investigate an architecture/paradigm/system, or create a knowledge document that goes beyond a terse summary.
Build or refactor React UI into a modular, reusable, themeable, Storybook-ready package. Use when asked to extract a widget into a component module/library, create theming APIs (CSS variables/parts), add slots or renderers, migrate styles into tokenized CSS, or produce Storybook stories that demonstrate default, themed, and unstyled variants.
Append-only Obsidian vault writing workflow for Manuel's vault at /home/manuel/code/wesen/go-go-golems/go-go-parc. Create new project notes, article/playbook-style knowledge notes, or new dated follow-up notes without overwriting historical notes unless the user explicitly asks. Use when the user mentions the Obsidian vault, asks to add a project description, wants a note modeled after an existing Projects/PROJ note such as PROJ - ZK Tool, wants a reusable ARTICLE/playbook/knowledge entry, or asks to include metadata such as the project date, repo directory, status, tags, and current purpose.
Print almanach pages on the AtomS3R thermal printer. Pair over BLE, print layouts from CLI or the remote rendering service at almanach.crib.scapegoat.dev.
Create exhaustive technology research and design packages for broad engineering topics. Use when the user asks to create a docmgr ticket for full-blown technology research/design, investigate unfamiliar libraries/protocols/frameworks, compare architecture options, gather source material, write intern-ready implementation guides, keep a diary, add small experiments, and optionally upload the final report to reMarkable.
Upload Markdown and source files to a reMarkable device as PDFs using the remarquee CLI (rmapi-backed). Use when the user asks to upload/send/export docs/examples to reMarkable, bundle multiple files into one PDF with a ToC, choose an /ai/YYYY/MM/DD destination, avoid or force overwrites, or troubleshoot pandoc/xelatex/rmapi auth.
| name | remarkable-render-pdf |
| description | Download a document from a reMarkable device/cloud via remarquee (rmapi-backed) and render it to an annotated PDF via `remarquee rmdoc render-v6`/`render-legacy`, optionally extracting the last N pages into a smaller PDF. Use when asked to export/render/download a reMarkable notebook/journal/document to PDF (full doc or page subset). |
remarquee status as a pre-check. Just run the command you need.remarquee cloud account before operations. Upload and cloud commands now auto-retry with reauth on 401/403. Pass --non-interactive and let the command handle auth automatically. If you see NOTE: auth expired, re-authenticating and retrying... in stderr, just let it proceed.cloud search instead of cloud ls + cloud ls when you know the document name — it's one call instead of two.qpdf and pdfinfoSearch by name (often easiest):
remarquee cloud search "journal #001" --match name --limit 50 --compact --non-interactive
If you already know the folder, list it:
remarquee cloud ls /Journals --long --non-interactive
Pick the remote path you want (example):
/Journals/#001 - 2025-12 - 2026-01-19
.rmdocoutdir="$HOME/remarkable_exports/$(date +%F)"
mkdir -p "$outdir"
remarquee cloud get "/Journals/#001 - 2025-12 - 2026-01-19" --out-dir "$outdir" --non-interactive
remarquee rmdoc inspect "$outdir/#001 - 2025-12 - 2026-01-19.rmdoc"
schema=cPages, render with render-v6render-legacyremarquee rmdoc render-v6 \
"$outdir/#001 - 2025-12 - 2026-01-19.rmdoc" \
--out "$outdir/#001 - 2025-12 - 2026-01-19-annotated.pdf" \
--force
Legacy fallback:
remarquee rmdoc render-legacy "$outdir/#001 - 2025-12 - 2026-01-19.rmdoc" --out "$outdir/#001 - 2025-12 - 2026-01-19-annotated.pdf" --force
Example: last 2 pages
pdf="$outdir/#001 - 2025-12 - 2026-01-19-annotated.pdf"
pages="$(pdfinfo "$pdf" | rg '^Pages:' | awk '{print $2}')"
start=$((pages-2+1))
qpdf --empty --pages "$pdf" "$start-$pages" -- "$outdir/#001 - 2025-12 - 2026-01-19-last-2-pages.pdf"
Use scripts/render_remote_to_pdf.py when you want a single deterministic command (download → render → optionally extract):
python3 scripts/render_remote_to_pdf.py \
--remote-path "/Journals/#001 - 2025-12 - 2026-01-19" \
--out-dir "$HOME/remarkable_exports/$(date +%F)" \
--extract-last 2 \
--force
It prints the generated paths on success.
If cloud tokens are missing and you need to authenticate, add --interactive.