@litfamily/litclaude 1.0.19 → 1.0.21
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +12 -0
- package/README.md +14 -8
- package/README_ko-KR.md +13 -8
- package/docs/assets/readme/badge-version.svg +1 -1
- package/docs/hooks.md +3 -2
- package/package.json +1 -1
- package/plugins/litclaude/.claude-plugin/plugin.json +1 -1
- package/plugins/litclaude/bin/litclaude-hook.js +4 -4
- package/plugins/litclaude/bin/litclaude-mcp.js +1 -1
- package/plugins/litclaude/lib/canonical-skill-resources.mjs +91 -31
- package/plugins/litclaude/skills/frontend-ui-ux/ATTRIBUTION.md +3 -6
- package/plugins/litclaude/skills/lit-docx/SKILL.md +50 -19
- package/plugins/litclaude/skills/lit-docx/references/components.md +207 -0
- package/plugins/litclaude/skills/lit-docx/references/direction-step.md +277 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/01-report-ko.md +166 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/02-brief-ko.md +105 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/03-manual-ko.md +164 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/04-proposal-ko.md +138 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/05-memo-en.md +55 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/06-journal-en.md +130 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/07-report-en.md +123 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/08-proposal-en.md +123 -0
- package/plugins/litclaude/skills/lit-docx/references/examples/09-memo-ko.md +45 -0
- package/plugins/litclaude/skills/lit-docx/references/page-composition.md +221 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/brief.md +143 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/journal.md +142 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/manual.md +148 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/memo.md +137 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/proposal.md +140 -0
- package/plugins/litclaude/skills/lit-docx/references/tonalities/report.md +156 -0
- package/plugins/litclaude/skills/lit-docx/scripts/convert_md_to_docx.py +274 -73
- package/plugins/litclaude/skills/lit-docx/scripts/docx_design.py +2449 -0
- package/plugins/litclaude/skills/lit-docx/scripts/docx_layout.py +959 -0
- package/plugins/litclaude/skills/lit-docx/scripts/qa_docx.py +66 -9
- package/plugins/litclaude/skills/lit-docx/scripts/slop_lint.py +18 -6
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/brief.yaml +28 -0
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/journal.yaml +25 -0
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/manual.yaml +26 -0
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/memo.yaml +26 -0
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/proposal.yaml +26 -0
- package/plugins/litclaude/skills/lit-docx/templates/tonalities/report.yaml +26 -0
- package/plugins/litclaude/skills/lit-pptx/SKILL.md +105 -48
- package/plugins/litclaude/skills/lit-pptx/references/agents/analyst.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/agents/thesis.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/anti-slop-checklist.md +59 -30
- package/plugins/litclaude/skills/lit-pptx/references/authoring-guide.md +119 -42
- package/plugins/litclaude/skills/lit-pptx/references/consensus-protocol.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/data-display-cookbook.md +13 -6
- package/plugins/litclaude/skills/lit-pptx/references/density-and-fill.md +164 -0
- package/plugins/litclaude/skills/lit-pptx/references/design-system-azure.md +7 -4
- package/plugins/litclaude/skills/lit-pptx/references/direction-step.md +186 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/01-pitch-signal-ko.md +210 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/02-pitch-studio-ko.md +147 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/03-business-review-ledger-ko.md +199 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/04-business-review-night-en.md +203 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/05-research-talk-paper-en.md +179 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/06-lecture-chalk-ko.md +199 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/07-data-review-night-en.md +194 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/08-image-led-atlas-ko.md +175 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/09-briefing-gazette-ko.md +204 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/10-status-update-ledger-ko.md +179 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/11-portfolio-studio-en.md +164 -0
- package/plugins/litclaude/skills/lit-pptx/references/examples/12-research-tutorial-chalk-en.md +174 -0
- package/plugins/litclaude/skills/lit-pptx/references/interview-protocol.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/layout-families.md +198 -0
- package/plugins/litclaude/skills/lit-pptx/references/stages/autopilot.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/stages/interview.md +2 -2
- package/plugins/litclaude/skills/lit-pptx/references/stages/plan.md +1 -1
- package/plugins/litclaude/skills/lit-pptx/references/title-treatments.md +109 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/atlas.md +139 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/chalk.md +148 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/gazette.md +149 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/ledger.md +149 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/night.md +147 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/paper.md +151 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/signal.md +150 -0
- package/plugins/litclaude/skills/lit-pptx/references/tonalities/studio.md +132 -0
- package/plugins/litclaude/skills/lit-pptx/scripts/compile-deck.js +64 -16
- package/plugins/litclaude/skills/lit-pptx/scripts/craft_extras.py +664 -3
- package/plugins/litclaude/skills/lit-pptx/scripts/deck_craft.py +48 -6
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/grid-resolver.js +521 -0
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/layout-resolver.js +362 -11
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/markdown-loader.js +6 -4
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/render-adapter-pptx.js +211 -65
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/render-pack.js +3559 -0
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/slide-ast.js +10 -6
- package/plugins/litclaude/skills/lit-pptx/scripts/lib/template-registry.js +286 -1
- package/plugins/litclaude/skills/lit-pptx/scripts/qa_deck.py +39 -10
- package/plugins/litclaude/skills/lit-pptx/scripts/validate_pptx.py +22 -7
- package/plugins/litclaude/skills/lit-pptx/templates/enrolled/AZURE-A2Z/layout-mapping.yaml +1 -1
- package/plugins/litclaude/skills/lit-pptx/templates/enrolled/AZURE-A2Z/template.yaml +4 -1
- package/plugins/litclaude/skills/lit-pptx/templates/enrolled/AZURE-PRO/layout-mapping.yaml +1 -1
- package/plugins/litclaude/skills/lit-pptx/templates/enrolled/AZURE-PRO/template.yaml +4 -1
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/atlas/pack.yaml +92 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/chalk/pack.yaml +97 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/gazette/pack.yaml +93 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/ledger/pack.yaml +94 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/night/pack.yaml +93 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/paper/pack.yaml +98 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/signal/pack.yaml +94 -0
- package/plugins/litclaude/skills/lit-pptx/templates/tonalities/studio/pack.yaml +92 -0
|
@@ -0,0 +1,147 @@
|
|
|
1
|
+
# Night
|
|
2
|
+
|
|
3
|
+
Night is a screen in a dark room. The ground is a deep blue-black, type is off-white, and the one
|
|
4
|
+
warm accent is amber, with a soft blue as the second series. Numbers sit in labelled rows at title
|
|
5
|
+
size at most, each with its basis, charts take most of the page, and a header band marks the slides that hold data. The
|
|
6
|
+
look keeps contrast high without glare, so a chart stays readable from the back of a hall. A plot or
|
|
7
|
+
diagram brought in as a picture on white would stand as a white card here: give it a `.dark` variant
|
|
8
|
+
beside the file, or its numbers as a `::: chart` block (authoring guide, "Charts").
|
|
9
|
+
|
|
10
|
+
## When to use it
|
|
11
|
+
|
|
12
|
+
Use Night for data and technical keynotes, product demos, architecture talks and quarterly reviews
|
|
13
|
+
shown on a big screen. It is the first candidate for a data-heavy review and moves up whenever the
|
|
14
|
+
brief says the deck is presented on a stage, at a demo day or in a darkened room. It also earns a
|
|
15
|
+
place when a third or more of the slides hold charts, because its families give charts the most
|
|
16
|
+
room of any pack.
|
|
17
|
+
|
|
18
|
+
## When it is the wrong choice
|
|
19
|
+
|
|
20
|
+
Night is wrong for decks that will be printed or read at a desk: a dark page wastes toner and tires
|
|
21
|
+
the eye on paper, so Ledger or Gazette serve those readers. It never mixes with light slides inside
|
|
22
|
+
one deck. It also lacks a table-led family, so a deck that is mostly lookup tables belongs in
|
|
23
|
+
Ledger.
|
|
24
|
+
|
|
25
|
+
## Tokens
|
|
26
|
+
|
|
27
|
+
These values are read from the pack file `pack.yaml` of the `night` tonality.
|
|
28
|
+
|
|
29
|
+
| Token | Value |
|
|
30
|
+
|---|---|
|
|
31
|
+
| ground / surface | `#0F1419` / `#1A222B` |
|
|
32
|
+
| ink / ink-muted | `#E8EDF2` / `#A3AEBA` |
|
|
33
|
+
| line | `#5A6672` |
|
|
34
|
+
| accent / accent-deep / accent-tint | `#E3A857` / `#F2C98A` / `#2A2418` |
|
|
35
|
+
| field / on-field | `#1E2A36` / `#F5F7FA` |
|
|
36
|
+
| positive / negative | `#6CC08B` / `#F08A7E` |
|
|
37
|
+
| series | `#E3A857`, `#7FB8E0`, `#A3AEBA`, `#C58FD0` |
|
|
38
|
+
| faces | display, title Pretendard Bold; body, label and numeral Pretendard Regular; `faces: a2z` swaps numerals to A2Z Light |
|
|
39
|
+
| ramp / hero | compact (body 13 pt, labels 11, sources 9) / not allowed, figures at most title size |
|
|
40
|
+
| density / variance | 10 / 6 |
|
|
41
|
+
| radius / edge | 6 / fill |
|
|
42
|
+
| table | header row on the surface colour with bold ink, hairline rows, no banding, numbers right-aligned, totals bold, the row the title names in the accent tint |
|
|
43
|
+
| chart | hairline gridlines, direct labels, accent on the item that matters and muted colours on the rest |
|
|
44
|
+
| decoration | header band, hairline rule, numeral, rail fill |
|
|
45
|
+
|
|
46
|
+
## Treatments by slide role
|
|
47
|
+
|
|
48
|
+
Text slides and image slides take `top-plain-large`, a title with no rule. Data slides take
|
|
49
|
+
`band`, the header band across the top. Charts with a takeaway take `side-rail`, sequences take
|
|
50
|
+
`kicker-numeral`, definitions take `side-rail`, and a pull quote takes `statement`. When the role's
|
|
51
|
+
treatment would leave an empty lower band, the engine tries the other treatments the family allows
|
|
52
|
+
and keeps the one that fills, while staying inside the variance cap. Write `title: side-rail` (or
|
|
53
|
+
another allowed treatment) under the `layout:` line to fix a slide's treatment; the business review
|
|
54
|
+
example does this for a big number and a timeline so the deck mixes enough compositions.
|
|
55
|
+
|
|
56
|
+
## Families and variants
|
|
57
|
+
|
|
58
|
+
- `full-chart`: one chart across the body with a short note under it.
|
|
59
|
+
- `chart-insight`: a chart with two or three takeaways.
|
|
60
|
+
- `kpi-row`: two to four headline values at title size, each with what it measures and its basis.
|
|
61
|
+
- `kpi-over-chart`: a value row above the trend it summarises.
|
|
62
|
+
- `dashboard-grid`: two to four small charts read together.
|
|
63
|
+
- `big-number`: a panel of two to four related figures (value, label, basis) with its evidence.
|
|
64
|
+
- `method`: how a metric is defined, formula first.
|
|
65
|
+
- `comparison`: two options or two periods, row by row.
|
|
66
|
+
- `timeline`: dated events on one axis.
|
|
67
|
+
- `statement`: a pull quote, or a display-led slide the user asks for.
|
|
68
|
+
|
|
69
|
+
Covers are `cover-numeral`, `cover-typographic` and `cover-figures`. Sections are `section-numeral`
|
|
70
|
+
and `section-field`. Closings are `closing-statement` and `closing-decision-box`.
|
|
71
|
+
|
|
72
|
+
Display slides. The typographic cover sets the year from the deck's date at title size beside the
|
|
73
|
+
topic name. A statement sets its quoted line in up to three balanced lines. A big-number panel
|
|
74
|
+
stands in amber on the dark amber tint, one row per figure with its label and basis. A closing's rows are keyed by date or amount, and the next step runs in the
|
|
75
|
+
field band above the footer.
|
|
76
|
+
|
|
77
|
+
## Do
|
|
78
|
+
|
|
79
|
+
Give each chart one highlighted series in amber and let the rest sit in muted tones. Keep titles
|
|
80
|
+
short labels (subject, measure, period): the open title and the band have tight glyph budgets, and the compile log warns when a title
|
|
81
|
+
runs over. Put the source on the last slide, inside the decision box. Use big numbers only for real
|
|
82
|
+
figures from the source, each with its basis and a sourced caption.
|
|
83
|
+
|
|
84
|
+
## Avoid
|
|
85
|
+
|
|
86
|
+
Avoid neon or saturated glows, a pure black ground, and any light slide mixed in. Avoid thin light
|
|
87
|
+
type below 18 pt; it breaks up on a projector. Do not lean on red and green alone for change; pair
|
|
88
|
+
the
|
|
89
|
+
colour with a sign and an arrow.
|
|
90
|
+
|
|
91
|
+
## Two worked slides
|
|
92
|
+
|
|
93
|
+
A value row above its trend states each number once with its basis, and the chart shows where it
|
|
94
|
+
came from and anchors the floor of the slide.
|
|
95
|
+
|
|
96
|
+
```markdown
|
|
97
|
+
---
|
|
98
|
+
layout: kpi-over-chart
|
|
99
|
+
|
|
100
|
+
## Cost per million requests, Q3 2025-Q3 2026
|
|
101
|
+
|
|
102
|
+
| Cost per million | Target | Change on Q2 |
|
|
103
|
+
|---|---|---|
|
|
104
|
+
| $1.82 | $2.00 | ▼ −9% |
|
|
105
|
+
| Q3 2026 | 9% under target | fourth fall in a row |
|
|
106
|
+
|
|
107
|
+
::: chart type=line unit="$"
|
|
108
|
+
| Quarter | Cost per million requests ($) |
|
|
109
|
+
|---|---|
|
|
110
|
+
| Q3 2025 | 2.61 |
|
|
111
|
+
| Q4 2025 | 2.34 |
|
|
112
|
+
| Q1 2026 | 2.12 |
|
|
113
|
+
| Q2 2026 | 2.00 |
|
|
114
|
+
| Q3 2026 | 1.82 |
|
|
115
|
+
> Cost per million requests by quarter (example data)
|
|
116
|
+
:::
|
|
117
|
+
---
|
|
118
|
+
```
|
|
119
|
+
|
|
120
|
+
A big-number panel holds three related figures with their basis and three lines of evidence, so
|
|
121
|
+
the figures are explained on the same page.
|
|
122
|
+
|
|
123
|
+
```markdown
|
|
124
|
+
---
|
|
125
|
+
layout: big-number
|
|
126
|
+
title: side-rail
|
|
127
|
+
|
|
128
|
+
## Incident recovery time, Q3 2026
|
|
129
|
+
|
|
130
|
+
| Median time to recover | Auto rollbacks | Slowest recovery |
|
|
131
|
+
|---|---|---|
|
|
132
|
+
| 8 min | 2 of 3 | 21 min |
|
|
133
|
+
| Q2: 27 min | before paging | database failover |
|
|
134
|
+
|
|
135
|
+
> Incident log, July-September 2026 (example)
|
|
136
|
+
|
|
137
|
+
- Recovery time is now under ten minutes
|
|
138
|
+
- The median fell from 27 minutes in Q2 after automatic rollback went live in July
|
|
139
|
+
- Two of the three incidents were rolled back before a person was paged
|
|
140
|
+
- The slowest recovery, 21 minutes, was a database failover in week 9
|
|
141
|
+
---
|
|
142
|
+
```
|
|
143
|
+
|
|
144
|
+
## Examples
|
|
145
|
+
|
|
146
|
+
`04-business-review-night-en.md` and `07-data-review-night-en.md` both use the pack defaults
|
|
147
|
+
(density 10, variance 6).
|
|
@@ -0,0 +1,151 @@
|
|
|
1
|
+
# Paper
|
|
2
|
+
|
|
3
|
+
Paper treats the screen like a printed journal page. The ground is white, the type is near-black,
|
|
4
|
+
and structure comes from thin rules rather than filled boxes. Figures do most of the talking, each
|
|
5
|
+
with a numbered caption and a source, and tables are set in the three-rule style readers know from
|
|
6
|
+
papers. One deep crimson is the only colour with a job: it marks the method or result the talk
|
|
7
|
+
proposes, and nothing else. The effect is calm and checkable, which is what a room of peers wants
|
|
8
|
+
when they are deciding whether to believe a result.
|
|
9
|
+
|
|
10
|
+
## When to use it
|
|
11
|
+
|
|
12
|
+
Pick Paper for research talks, thesis defences, lab meetings and technical reviews where the
|
|
13
|
+
audience reads the evidence closely. It is the first candidate for the research-talk deck type, and
|
|
14
|
+
it moves up the list when the source carries numbered citations, DOIs or a references section, or
|
|
15
|
+
when a third or more of the slides hold a chart. It also suits a written analysis that happens to be
|
|
16
|
+
delivered as slides, because its hairline tables and numbered figures survive printing.
|
|
17
|
+
|
|
18
|
+
## When it is the wrong choice
|
|
19
|
+
|
|
20
|
+
Paper reads as cold to a sales, launch or investor audience; there Signal or Studio carry the
|
|
21
|
+
energy the room expects. It is also a poor fit for a deck of mostly photographs, since the pack has
|
|
22
|
+
no full-bleed image family and keeps pictures inside the grid. If the talk is given on a large
|
|
23
|
+
stage in a dark room, Night keeps charts legible at distance where Paper's thin rules disappear.
|
|
24
|
+
|
|
25
|
+
## Tokens
|
|
26
|
+
|
|
27
|
+
These values are read from the pack file `pack.yaml` of the `paper` tonality.
|
|
28
|
+
|
|
29
|
+
| Token | Value |
|
|
30
|
+
|---|---|
|
|
31
|
+
| ground / surface | `#FFFFFF` / `#F5F5F6` |
|
|
32
|
+
| ink / ink-muted | `#111418` / `#50565E` |
|
|
33
|
+
| line | `#C9CDD2` |
|
|
34
|
+
| accent / accent-deep / accent-tint | `#8C1C2B` / `#5E121C` / `#F6E6E8` |
|
|
35
|
+
| field / on-field | `#111418` / `#FFFFFF` |
|
|
36
|
+
| positive / negative | `#1E6B3A` / `#50565E` |
|
|
37
|
+
| series | `#8C1C2B`, `#111418`, `#50565E`, `#2E5E8A` |
|
|
38
|
+
| faces | display and title Pretendard Bold; body, label and numeral Pretendard Regular (`faces: a2z` keeps the same map) |
|
|
39
|
+
| ramp / hero | compact (body 13 pt, labels 11, sources 9) / no hero step, figures at most title size |
|
|
40
|
+
| density / variance | 10 / 5 |
|
|
41
|
+
| radius / edge | 0 / border |
|
|
42
|
+
| table | booktabs: heavy rule over the header, light rule under it, heavy rule at the end; no fill, no banding; numbers right-aligned; totals bold |
|
|
43
|
+
| chart | hairline gridlines, direct labels, series in order (the proposed method takes the accent, baselines take ink and muted ink) |
|
|
44
|
+
| decoration | hairline rule, box outline |
|
|
45
|
+
|
|
46
|
+
Negative values share the muted ink on purpose: a drop is shown with a sign and a glyph, never with
|
|
47
|
+
a
|
|
48
|
+
warning colour that would compete with the accent.
|
|
49
|
+
|
|
50
|
+
## Treatments by slide role
|
|
51
|
+
|
|
52
|
+
Text slides, data tables, sequences and reference pages take `top-rule`, where the rule under the
|
|
53
|
+
title is a full hairline. Charts and figures that carry the slide take `bottom-anchor`, so the
|
|
54
|
+
picture fills the upper body and the title labels it from below. Definitions and method slides
|
|
55
|
+
take `side-rail`, a comparison is drawn as a ruled table under a `top-rule` title (the pack's
|
|
56
|
+
`structure`, kept even when the deck already leans on that title), and a quoted line takes `statement`. The engine keeps the role's treatment when
|
|
57
|
+
the body fills to the density band; when it would leave a tall empty band, it lays the slide out
|
|
58
|
+
under the other treatments the family allows and keeps the one that fills best, while capping how
|
|
59
|
+
many treatments one deck mixes. A slide can ask for a specific treatment with a `title:` line under
|
|
60
|
+
its `layout:` line, for example `title: top-rule`, as long as the family allows it.
|
|
61
|
+
|
|
62
|
+
## Families and variants
|
|
63
|
+
|
|
64
|
+
- `figure-academic`: one figure with its numbered caption and a short observation, limitation and
|
|
65
|
+
implication.
|
|
66
|
+
- `figure-pair`: two figures side by side, for before and after or two conditions.
|
|
67
|
+
- `method`: a formula or process figure beside the definitions of its symbols.
|
|
68
|
+
- `chart-insight`: a native chart with two or three takeaways beside it.
|
|
69
|
+
- `full-chart`: a chart across the body when the plot itself is the argument.
|
|
70
|
+
- `table-insight`: a booktabs table with its caption and the takeaway it supports.
|
|
71
|
+
- `comparison`: contributions against limits, or two methods row by row.
|
|
72
|
+
- `quote`: a participant's or reviewer's words as a statement.
|
|
73
|
+
- `statement`: an epigraph or quoted line; the research question and the finding go in a labelled
|
|
74
|
+
content slide with their evidence.
|
|
75
|
+
- `text-column`: running argument at reading measure.
|
|
76
|
+
- `references-appendix`: the reference list, one entry per row.
|
|
77
|
+
|
|
78
|
+
Covers are `cover-index` (lists the deck's sections, so it needs at least two), `cover-typographic`
|
|
79
|
+
and `cover-split-image`. Sections are `section-rule` and `section-field`. Closings are
|
|
80
|
+
`closing-summary-list` and `closing-ask`.
|
|
81
|
+
|
|
82
|
+
Display slides. The typographic cover is a journal title page: a heavy rule, the title, a hairline,
|
|
83
|
+
then the presenter and, under it, the place and date. A statement sits between two hairlines, like
|
|
84
|
+
an epigraph. A section shows the deck's parts in one line under its rule, this part in the accent. A
|
|
85
|
+
closing's next step sits under a rule at the foot of the page, its rows keyed by date.
|
|
86
|
+
|
|
87
|
+
## Do
|
|
88
|
+
|
|
89
|
+
Number every figure and table caption and give each a source. Title each slide with what it shows
|
|
90
|
+
("Conversion at 60 min by feed rate"), and put the finding in the first observation. Keep the crimson for the proposed
|
|
91
|
+
method in every chart, so the audience learns its meaning on the first plot. Put the limitation on
|
|
92
|
+
the same slide as the result it limits. End with the reference list, so the last page carries the
|
|
93
|
+
deck's sources.
|
|
94
|
+
|
|
95
|
+
## Avoid
|
|
96
|
+
|
|
97
|
+
Avoid the short accent bar under a title; in this direction the rule is a hairline. Avoid coloured
|
|
98
|
+
card fills, which turn a paper into a dashboard. Do not let the accent mark anything that is not the
|
|
99
|
+
proposed method, and never show a figure without its numbered caption.
|
|
100
|
+
|
|
101
|
+
## Two worked slides
|
|
102
|
+
|
|
103
|
+
A table slide labels the table in the title and states the takeaway first beside it, so the body
|
|
104
|
+
fills and the reader does not have to compute the gain.
|
|
105
|
+
|
|
106
|
+
```markdown
|
|
107
|
+
---
|
|
108
|
+
layout: table-insight
|
|
109
|
+
|
|
110
|
+
## Conversion at 60 min by feed rate and feed mode
|
|
111
|
+
|
|
112
|
+
| Feed rate (L/min) | Single feed (%) | Staged feed (%) | Gain (points) |
|
|
113
|
+
|---|---|---|---|
|
|
114
|
+
| 0.5 | 52 | 77 | 25 |
|
|
115
|
+
| 0.75 | 51 | 76 | 25 |
|
|
116
|
+
| 1.0 | 50 | 75 | 25 |
|
|
117
|
+
| 1.5 | 47 | 71 | 24 |
|
|
118
|
+
| 2.0 | 41 | 64 | 23 |
|
|
119
|
+
|
|
120
|
+
> Table 1. Conversion at 60 minutes by feed rate, three replicates each (example data)
|
|
121
|
+
|
|
122
|
+
- The gain holds at all five feed rates tested (+23 to +25 points)
|
|
123
|
+
- The relative gain is 48-56% at every rate
|
|
124
|
+
- The gap narrows by two points at the highest rate
|
|
125
|
+
- Heat removal, not the feed schedule, limits the fastest runs
|
|
126
|
+
---
|
|
127
|
+
```
|
|
128
|
+
|
|
129
|
+
A method slide pairs the process figure with every symbol the talk will use, so later charts need
|
|
130
|
+
no legend.
|
|
131
|
+
|
|
132
|
+
```markdown
|
|
133
|
+
---
|
|
134
|
+
layout: method
|
|
135
|
+
|
|
136
|
+
## Three-stage feed layout and symbols
|
|
137
|
+
|
|
138
|
+

|
|
139
|
+
|
|
140
|
+
- **Layout** feed split into three stages along the reactor
|
|
141
|
+
- **X** conversion, the fraction of feed that reacted
|
|
142
|
+
- **k** first-order rate constant, 1/min
|
|
143
|
+
- **τ** residence time per stage, min
|
|
144
|
+
- **n** number of feed stages, three in every run
|
|
145
|
+
- **q** feed split between stages, 50/30/20 by volume
|
|
146
|
+
---
|
|
147
|
+
```
|
|
148
|
+
|
|
149
|
+
## Examples
|
|
150
|
+
|
|
151
|
+
`05-research-talk-paper-en.md` uses the pack defaults (density 10, variance 5).
|
|
@@ -0,0 +1,150 @@
|
|
|
1
|
+
# Signal
|
|
2
|
+
|
|
3
|
+
Signal is the direction for a talk given out loud. Each slide carries one topic with the facts that
|
|
4
|
+
support it, set on the reading ramp so the room can follow while the speaker talks and a reader can
|
|
5
|
+
still check the figures later. The ground stays white, almost everything is near-black ink, and a
|
|
6
|
+
single red-orange accent marks the one figure or word the slide exists for. There are no cards and
|
|
7
|
+
very few lines; space, weight and colour do the grouping.
|
|
8
|
+
|
|
9
|
+
## When to use it
|
|
10
|
+
|
|
11
|
+
Reach for Signal when someone will stand in front of decision makers and speak for ten to fifteen
|
|
12
|
+
minutes: an investment pitch, a product launch, a proposal that has to land in one sitting. The
|
|
13
|
+
content signals that point here are short bodies (about 40 Korean glyphs or 80 Latin characters per
|
|
14
|
+
content slide or less), a handful of real headline numbers, and a story that moves from a problem
|
|
15
|
+
to an ask. It is the first candidate for the pitch deck type, and the third one for an image-led
|
|
16
|
+
launch when the pictures are few.
|
|
17
|
+
|
|
18
|
+
## When it is the wrong choice
|
|
19
|
+
|
|
20
|
+
Signal fails when the deck will be read alone later, because its open pages hold less detail than a
|
|
21
|
+
reader needs without a speaker. It also fails when tables make up a third of the slides or
|
|
22
|
+
more: tables at Signal's sizes either run out of room or shrink to a size that contradicts the rest
|
|
23
|
+
of the deck. If the brief says the deck is sent ahead or circulated, Ledger or Gazette fits better.
|
|
24
|
+
|
|
25
|
+
## Tokens
|
|
26
|
+
|
|
27
|
+
The values below come from the pack file and are what the engine draws.
|
|
28
|
+
|
|
29
|
+
| Token | Value |
|
|
30
|
+
|---|---|
|
|
31
|
+
| ground / surface | #FFFFFF / #F4F4F5 |
|
|
32
|
+
| ink / ink-muted / line | #15131A / #55525C / #D4D4D8 |
|
|
33
|
+
| accent / accent-deep / accent-tint | #C8361A / #6E1A0C / #FBE7E2 |
|
|
34
|
+
| field / on-field | #C8361A / #FFFFFF |
|
|
35
|
+
| positive / negative | #1E7A3F / #15131A |
|
|
36
|
+
| series | #C8361A, #15131A, #6B6870, #2F6F8F |
|
|
37
|
+
| faces | Pretendard Bold for display, title and numerals, Pretendard Regular for body and labels; `faces: a2z` swaps to A2Z Black display and numerals, A2Z Bold titles |
|
|
38
|
+
| ramp / hero | compact (body 13 pt, labels 11, sources 9) / not allowed (figures at most title size, 26 pt) |
|
|
39
|
+
| density / variance | 10 / 6 |
|
|
40
|
+
| radius / edge | 0 / fill |
|
|
41
|
+
| table | open style, no header fill, rules at top and bottom, protagonist row in accent-tint, bold totals |
|
|
42
|
+
| chart | no gridlines, direct labels, accent on the highlighted bar over muted bars |
|
|
43
|
+
|
|
44
|
+
Density 10 sets the pitch on the compact grid (24 pt side margins), body 13 pt and no step-up: a
|
|
45
|
+
pitch read after the meeting carries its basis and sources on every slide. Lower it to 8 (body 14 pt,
|
|
46
|
+
36 pt margins) for a pitch given only from the stage. Losses use ink with a ▼ sign, because red already means
|
|
47
|
+
"look here" in this direction.
|
|
48
|
+
|
|
49
|
+
## Treatments by slide role
|
|
50
|
+
|
|
51
|
+
Content slides take `top-plain-large`, data and sequence slides `top-rule`, data with a takeaway
|
|
52
|
+
`top-plain-large`, image slides `bottom-anchor`, statements `statement`, and references and
|
|
53
|
+
definitions `top-rule`. The engine treats that map as a starting point: for each slide it lays out
|
|
54
|
+
every treatment the family allows, keeps the role's own when the body fills, and otherwise picks the
|
|
55
|
+
one that leaves the least empty band without letting one title cover more than about a third of the
|
|
56
|
+
deck. To force a treatment, put `title: bottom-anchor` (or another allowed id) on the line after
|
|
57
|
+
`layout:`.
|
|
58
|
+
|
|
59
|
+
## Families and variants
|
|
60
|
+
|
|
61
|
+
- `statement` is kept for a pull quote or a display-led deck the user asks for; a claim is not a title.
|
|
62
|
+
- `big-number` is a data panel of two to four related figures, each with its label and basis, beside
|
|
63
|
+
the facts that explain them.
|
|
64
|
+
- `image-split` pairs a product picture with three short labelled points.
|
|
65
|
+
- `asymmetric-feature` gives a picture or number the narrow side and the explanation the wide one.
|
|
66
|
+
- `process` shows three to five steps across the page, each with a head and two lines of detail.
|
|
67
|
+
- `kpi-row` runs two to four results at title size, each with its label and a basis row under it.
|
|
68
|
+
- `chart-insight` and `full-chart` carry growth curves and comparisons with one highlighted series.
|
|
69
|
+
- `comparison` reads two columns across shared rows, such as before and after.
|
|
70
|
+
- `table-insight` holds the one table a pitch needs, usually unit economics, with its takeaway.
|
|
71
|
+
- `quote` sets a customer's sentence as the statement.
|
|
72
|
+
|
|
73
|
+
Covers are `cover-numeral`, `cover-typographic` and `cover-split-field`; sections are
|
|
74
|
+
`section-field` and `section-numeral`; closings are `closing-statement` and `closing-ask`.
|
|
75
|
+
|
|
76
|
+
Display slides. The typographic cover is the red field from edge to edge, the topic name in up to
|
|
77
|
+
three balanced lines at display size, anchored low. A statement (pull quote) sets its text in up to
|
|
78
|
+
three balanced lines on white. A big number is a red data panel: one row per figure with value, label
|
|
79
|
+
and basis in white, its evidence beside it. A closing's rows carry their amounts at title size, each
|
|
80
|
+
with a bar for its share of the ask, and the next step runs in
|
|
81
|
+
a red band across the grid just above the footer.
|
|
82
|
+
|
|
83
|
+
## Do
|
|
84
|
+
|
|
85
|
+
- Write every title as a topic label (subject, measure, period); the claim is the first body line.
|
|
86
|
+
- Give every figure its basis or comparison and a caption with period and source.
|
|
87
|
+
- Give a big-number panel two to four figures and two or three lines of evidence.
|
|
88
|
+
- Keep bullet lists to four items or fewer and let each item carry a number or a name.
|
|
89
|
+
- End on `closing-ask` with the amount, its uses in a table and the next meeting with an owner.
|
|
90
|
+
|
|
91
|
+
## Avoid
|
|
92
|
+
|
|
93
|
+
- Red for a loss; red belongs to the protagonist.
|
|
94
|
+
- Display text running past two lines, or one giant figure alone on a slide.
|
|
95
|
+
- Cards or boxed groups; this direction groups with space.
|
|
96
|
+
- A table slide with no sentence that tells the room what to take from it.
|
|
97
|
+
|
|
98
|
+
## Two worked slides
|
|
99
|
+
|
|
100
|
+
A big number on its own reads as decoration, so the panel holds three related figures with their
|
|
101
|
+
basis, and the evidence sits beside it and ends level with it.
|
|
102
|
+
|
|
103
|
+
```markdown
|
|
104
|
+
---
|
|
105
|
+
layout: big-number
|
|
106
|
+
|
|
107
|
+
## 수도권 세탁 대행 시장 규모와 앱 주문 비중
|
|
108
|
+
|
|
109
|
+
| 시장 규모 (연) | 앱 주문 비중 | 수거·배송 하는 세탁소 |
|
|
110
|
+
|---|---|---|
|
|
111
|
+
| 1조 2천억 원 | 6% | 20% |
|
|
112
|
+
| 2025년 기준 | 2023년 3% | 다섯 곳 중 한 곳 |
|
|
113
|
+
|
|
114
|
+
> 수도권 세탁 대행 시장 추정, 예시 업계 통계 2025 (예시)
|
|
115
|
+
|
|
116
|
+
- 시장은 아직 대부분 오프라인이다: 앱 주문은 6%, 약 720억 원
|
|
117
|
+
- 동네 세탁소 다섯 곳 중 네 곳은 수거·배송을 하지 않는다
|
|
118
|
+
- 앱 주문 비중은 2023년 3%에서 2025년 6%로 늘었다
|
|
119
|
+
---
|
|
120
|
+
```
|
|
121
|
+
|
|
122
|
+
The closing labels the request in the title, shows where the money goes, states the ask in the
|
|
123
|
+
first line and ends with the next step.
|
|
124
|
+
|
|
125
|
+
```markdown
|
|
126
|
+
---
|
|
127
|
+
layout: closing-ask
|
|
128
|
+
|
|
129
|
+
## 18개월 운영 자금 30억 원 사용 계획
|
|
130
|
+
|
|
131
|
+
| 쓰임 | 금액 | 시점 |
|
|
132
|
+
|---|---|---|
|
|
133
|
+
| 수거·배송 권역 확대 (6개 구 → 15개 구) | 14억 원 | 2027년 상반기 |
|
|
134
|
+
| 제휴 세탁소 품질 관리 체계 | 6억 원 | 2026년 4분기 |
|
|
135
|
+
| 앱 개발 인력 | 5억 원 | 상시 |
|
|
136
|
+
| 신규 제휴 세탁소 120곳 모집 | 3억 원 | 2027년 1분기 |
|
|
137
|
+
| 운영 예비비 | 2억 원 | 필요 시 |
|
|
138
|
+
| 합계 | 30억 원 | 18개월 |
|
|
139
|
+
|
|
140
|
+
> 표 2. 투자금 쓰임 (예시 데이터)
|
|
141
|
+
|
|
142
|
+
- 요청: 18개월 운영 자금 30억 원 투자 승인
|
|
143
|
+
- 기대 성과: 18개월 뒤 월 주문 6만 건, 월 영업이익 흑자 전환
|
|
144
|
+
- 다음 단계: 11월 둘째 주 실사 미팅, 담당 예시 재무팀
|
|
145
|
+
---
|
|
146
|
+
```
|
|
147
|
+
|
|
148
|
+
## Examples
|
|
149
|
+
|
|
150
|
+
`01-pitch-signal-ko.md`
|
|
@@ -0,0 +1,132 @@
|
|
|
1
|
+
# Studio
|
|
2
|
+
|
|
3
|
+
Studio treats a deck like a magazine spread. The twelve-column grid shows through as thin column
|
|
4
|
+
lines, titles often stand in a side rail or open the page at title size, real figures sit in labelled
|
|
5
|
+
panels with their basis, and pages split unevenly so that one side clearly leads. A dark ochre accent marks the single
|
|
6
|
+
thing to look at; everything else is black, grey and white.
|
|
7
|
+
|
|
8
|
+
## When to use it
|
|
9
|
+
|
|
10
|
+
Studio fits stories that are told more than they are audited: a design or brand narrative, a
|
|
11
|
+
strategy told through cases, a public lecture, a portfolio of built work, or a pitch for a product
|
|
12
|
+
people judge by seeing it. The content signals that move Studio up are a high share of slides with
|
|
13
|
+
pictures (about 30% or more) and short to medium bodies. It is the second candidate for pitches,
|
|
14
|
+
image-led decks and public lectures, and it can lead when the pictures are strong but the narrative
|
|
15
|
+
matters as much as the images.
|
|
16
|
+
|
|
17
|
+
## When it is the wrong choice
|
|
18
|
+
|
|
19
|
+
The editorial splits starve table-heavy data. A business review with several tables, a status
|
|
20
|
+
update or a briefing that is read at a desk all need more rows on each page than Studio's uneven
|
|
21
|
+
splits give them; Ledger or Gazette serve those. Studio also needs real figures for its numeral
|
|
22
|
+
panels: a deck without counts, shares or years loses one of the direction's main devices.
|
|
23
|
+
|
|
24
|
+
## Tokens
|
|
25
|
+
|
|
26
|
+
Read from the pack file; these are the values the engine draws.
|
|
27
|
+
|
|
28
|
+
| Token | Value |
|
|
29
|
+
|---|---|
|
|
30
|
+
| ground / surface | #FFFFFF / #F2F2F0 |
|
|
31
|
+
| ink / ink-muted / line | #141414 / #525252 / #D2D2CE |
|
|
32
|
+
| accent / accent-deep / accent-tint | #8A5A00 / #4D3200 / #F5ECD9 |
|
|
33
|
+
| field / on-field | #141414 / #FFFFFF |
|
|
34
|
+
| positive / negative | #1E7A3F / #B3261E |
|
|
35
|
+
| series | #8A5A00, #141414, #525252, #3B5F7F |
|
|
36
|
+
| faces | Pretendard Bold for display, title and numerals, Pretendard Regular for body and labels; `faces: a2z` swaps to A2Z Black display and numerals, A2Z Medium titles |
|
|
37
|
+
| ramp / hero | compact (body 13 pt, labels 11, sources 9) / not allowed, figures at most title size |
|
|
38
|
+
| density / variance | 10 / 8 |
|
|
39
|
+
| radius / edge | 0 / fill |
|
|
40
|
+
| table | open style, no header fill, a 2 pt ink rule under the header, no row rules, bold totals |
|
|
41
|
+
| chart | no gridlines, direct labels, accent on the highlighted series over muted ones |
|
|
42
|
+
|
|
43
|
+
Density 10 puts the deck on the compact grid with body 13 pt. Variance 8 lets one deck mix up to five
|
|
44
|
+
title treatments, which is the point of this direction.
|
|
45
|
+
|
|
46
|
+
## Treatments by slide role
|
|
47
|
+
|
|
48
|
+
Content, data with a takeaway, references, definitions and sequences take `side-rail`; data
|
|
49
|
+
slides take `top-plain-large`; a pull quote takes `statement`; image slides take `side-rail`, the
|
|
50
|
+
picture across columns 5-12 and its facts in the rail under the title (an asymmetric split, where
|
|
51
|
+
Atlas would set the title under the picture); the `image-full` family takes `overlay`. For each slide the engine tries every
|
|
52
|
+
treatment its family
|
|
53
|
+
allows and keeps the role default when the body fills, otherwise the one that leaves the least empty
|
|
54
|
+
band, while keeping any single treatment under about a third of the deck. A line such as
|
|
55
|
+
`title: top-plain-large` after `layout:` overrides that choice for one slide.
|
|
56
|
+
|
|
57
|
+
## Families and variants
|
|
58
|
+
|
|
59
|
+
- `asymmetric-feature` gives a tall picture the narrow side and four lines of argument the wide one.
|
|
60
|
+
- `big-number` sets two to four real figures as a panel (value, label, basis) with the facts that
|
|
61
|
+
explain them.
|
|
62
|
+
- `image-split` and `image-full` carry the work itself; the full-bleed slide puts its title on a
|
|
63
|
+
panel.
|
|
64
|
+
- `photo-grid` shows two or three pictures side by side with one shared sentence.
|
|
65
|
+
- `text-two-column` holds a brief and a response, or two halves of an argument.
|
|
66
|
+
- `chart-insight` and `kpi-row` carry the few numbers the story needs.
|
|
67
|
+
- `timeline` shows a programme or a history along one axis.
|
|
68
|
+
- `comparison` reads before and after across shared rows.
|
|
69
|
+
|
|
70
|
+
Covers are `cover-split-image` (which takes the deck's first picture), `cover-rail` and
|
|
71
|
+
`cover-numeral`; sections are `section-numeral` and `section-rail`; closings are
|
|
72
|
+
`closing-contact-split`
|
|
73
|
+
and `closing-summary-list`.
|
|
74
|
+
|
|
75
|
+
Display slides. The cover opens on the deck's first picture when there is one, otherwise on the
|
|
76
|
+
rail. A statement stands between two hairlines on the open page (a light rail beside it stood empty). A
|
|
77
|
+
big-number panel stands in ochre on its tint, its figures at title size. The contact-split closing keeps the field: it holds each
|
|
78
|
+
row's amount or date in line with its row, and the next step runs under the rows on the white side.
|
|
79
|
+
|
|
80
|
+
## Do
|
|
81
|
+
|
|
82
|
+
- Let the picture or the number lead each slide and keep the text to the side that serves it.
|
|
83
|
+
- Use a numeral only for a real figure (a count, a share, a year), with its label and basis.
|
|
84
|
+
- Title each slide with its topic; the claim opens the body beside the picture.
|
|
85
|
+
- Alternate splits: after a side-rail slide, follow with a top title or a full-bleed picture.
|
|
86
|
+
- Keep titles short; a side-rail title has about eight Korean glyphs or sixteen Latin characters per
|
|
87
|
+
line, and a long one overflows its frame.
|
|
88
|
+
|
|
89
|
+
## Avoid
|
|
90
|
+
|
|
91
|
+
- Numerals used as ornaments, such as step numbers on points that are not a sequence.
|
|
92
|
+
- Column lines running through text; they belong in gutters.
|
|
93
|
+
- Two side-rail slides with the same split back to back.
|
|
94
|
+
- Centred body text; Studio aligns everything to the column lines.
|
|
95
|
+
|
|
96
|
+
## Two worked slides
|
|
97
|
+
|
|
98
|
+
A tall portrait picture takes the narrow side, so the five facts beside it fill the page; the first
|
|
99
|
+
carries the claim.
|
|
100
|
+
|
|
101
|
+
```markdown
|
|
102
|
+
---
|
|
103
|
+
layout: asymmetric-feature
|
|
104
|
+
|
|
105
|
+
## 1인 가구 이사 주기와 가구 폐기 실태
|
|
106
|
+
|
|
107
|
+

|
|
108
|
+
|
|
109
|
+
- 1인 가구는 2년마다 이사하며 가구를 버린다
|
|
110
|
+
- 수도권 1인 가구의 평균 거주 기간은 2.1년이다 (예시 조사)
|
|
111
|
+
- 이사 때 버려지는 가구의 절반은 산 지 3년이 안 된 것이다
|
|
112
|
+
- 대형 폐기물 수수료와 운반비로 한 번에 평균 9만 원을 쓴다
|
|
113
|
+
- 새 집 구조에 맞지 않아 버리는 경우가 가장 많다
|
|
114
|
+
---
|
|
115
|
+
```
|
|
116
|
+
|
|
117
|
+
A full-bleed picture carries its label title on a panel; the caption, which states what changed,
|
|
118
|
+
stays on the panel with it.
|
|
119
|
+
|
|
120
|
+
```markdown
|
|
121
|
+
---
|
|
122
|
+
layout: image-full
|
|
123
|
+
|
|
124
|
+
## East branch reading room after the renovation
|
|
125
|
+
|
|
126
|
+

|
|
127
|
+
---
|
|
128
|
+
```
|
|
129
|
+
|
|
130
|
+
## Examples
|
|
131
|
+
|
|
132
|
+
`02-pitch-studio-ko.md`, `11-portfolio-studio-en.md`
|