agentleFS
fmindfmind/dot/skills/draft-typst-deck/SKILL.md

draft-typst-deck

Skill13 starsChanged yesterdayMITView on GitHub

Create 16:9 Typst slides, branded visual cards, PDF decks, and private Google Slides via gws.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Editable-shape Google Slides: when stakeholders must edit text boxes directly in Slides, adapt editable_deck.py (Google Sans native shapes …

    line 22
  2. Very specific:

    Max 30 visible words, zero prose backticks: use #hl[...] or #pill(...) instead of code styling on non-code words, enforce --max-words 30 in …

    line 28
  3. Very specific:

    Run build_typst_deck.py with isolated stdlib Python (python3 -I):

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Draft Typst Slide Deck

Build crisp 16:9 (960pt x 540pt) Typst slide decks and standalone branded visual cards with editable .typ sources, compiled PDF/PNG/PPTX outputs, speaker notes, and optional Google Drive / Google Slides publishing via gws. Default to the Fmind Theme (~/.agents/DESIGN.md: #FFFFFF canvas, #174EA6 accent, Google Sans), or substitute a client's brand tokens and fonts (never mix brands).

Format Choice (Typst Deck vs Editable Slides vs Visual Card)

Apply humanize (outlines, primitives) on every deck:

  1. Typst PDF / Retina PPTX -> Private Google Slides (default): adapt fmind-slides.typ + deck.typ, add #speaker-note("..."), compile with check --pptx, and publish via upload --google-slides --source-dir .. Drive converts the .pptx privately (Retina images + editable speaker notes) without public URLs.
  2. Editable-shape Google Slides: when stakeholders must edit text boxes directly in Slides, adapt editable_deck.py (Google Sans native shapes + 2x Typst PNG cards via uv run --script), or patch an existing template via gws slides presentations batchUpdate (replaceAllText), and upload with --google-slides --source-dir ..
  3. Standalone branded PNG/SVG card: adapt figure-card.typ (#show: brand-card.with(width: 640pt)) or compile figures/*.typ with render (or check --render-figures) for Docs (draft-doc), reports, email, or web.

Slide Restraint Rules (Zero Slide Dumps)

  • One idea, one focal centerpiece: pair an action-led headline with one visual (kpi-card strip, bar-panel, split-cards, or step-card flow). Never stack two heavy visuals on one slide.
  • Max 30 visible words, zero prose backticks: use #hl[...] or #pill(...) instead of code styling on non-code words, enforce --max-words 30 in check, and move caveats and formulas to #speaker-note("...").
  • Readable minimums & escaping: titles 26 pt, cards 15–16 pt, chart labels 12+ pt; follow charts-and-diagrams.md ($, ~, * differ between "..." and [...]).

Workflow (check, render, inspect, upload)

Run build_typst_deck.py with isolated stdlib Python (python3 -I):

  1. Validate deck, enforce word ceiling, export PNG/PDF/PPTX:

    python3 -I ~/.agents/skills/draft-typst-deck/scripts/build_typst_deck.py check deck.typ \
      --expected-slides 4 --max-words 30 --pptx --json

    Pass --render-figures --figure-format png to also compile figures/*.typ, or compile one standalone card directly:

    python3 -I ~/.agents/skills/draft-typst-deck/scripts/build_typst_deck.py render figure-card.typ --json

    Fails closed on bad Typst escapes, empty *bold* stars, missing fonts, word-limit overruns, or slide overflow.

  2. Inspect every PNG (_preview/<stem>-*.png or card output) for spacing, contrast, and calm per verification-and-drive.md.
  3. Dry-run and publish to Drive / Google Slides (after explicit approval):

    python3 -I ~/.agents/skills/draft-typst-deck/scripts/build_typst_deck.py upload deck.pptx \
      --drive-folder "<FOLDER_ID>" --google-slides --source-dir . --dry-run --json

    Rerun without --dry-run (or pass --file-id "<ID>" to update in place); --source-dir . syncs .typ and root .py sources into Drive assets/<slug>/. Verify remote integrity before finishing.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.