Skip to content

AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He

docs(ppt): clarify canvas dimensions

hugohe3cbc209a6caa1
diff --git a/skills/ppt-master/SKILL.md b/skills/ppt-master/SKILL.md
index a38062af..610ec2f9 100644
--- a/skills/ppt-master/SKILL.md
+++ b/skills/ppt-master/SKILL.md
@@ -184,7 +184,7 @@ When the user provides non-Markdown content, convert immediately:
python3 ${SKILL_DIR}/scripts/project_manager.py init <project_name> --format <format>
```
-Format options: `ppt169` (default), `ppt43`, `xhs`, `story`, etc. For the full format list, see `references/canvas-formats.md`.
+Format options must be named with concrete dimensions. Default: `ppt169` = `1280x720`, `viewBox="0 0 1280 720"`. Other examples: `ppt43` = `1024x768`, `story` = `1080x1920`, `banner` = `1920x1080`. For the full format list, see `references/canvas-formats.md`.
Import source content (choose based on the situation):
diff --git a/skills/ppt-master/references/canvas-formats.md b/skills/ppt-master/references/canvas-formats.md
index c5863f21..0be72bf4 100644
--- a/skills/ppt-master/references/canvas-formats.md
+++ b/skills/ppt-master/references/canvas-formats.md
@@ -4,17 +4,18 @@
## Format Quick Reference
-| Format | viewBox | Ratio | Use Case |
-|--------|---------|-------|----------|
-| PPT 16:9 | `0 0 1280 720` | 16:9 | Business presentations, meetings |
-| PPT 4:3 | `0 0 1024 768` | 4:3 | Traditional projectors, academic talks |
-| Xiaohongshu (RED) | `0 0 1242 1660` | 3:4 | Image-text sharing, knowledge posts |
-| WeChat Moments / IG | `0 0 1080 1080` | 1:1 | Square posters, brand showcases |
-| Story / TikTok | `0 0 1080 1920` | 9:16 | Vertical stories, short video covers |
-| WeChat Article Header | `0 0 900 383` | 2.35:1 | WeChat article cover images |
-| Landscape Banner | `0 0 1920 1080` | 16:9 | Web banners, digital screens |
-| Portrait Poster | `0 0 1080 1920` | 9:16 | Phone screens, elevator ads |
-| A4 Print | `0 0 1240 1754` | 1:sqrt(2) | Print posters, flyers |
+| ID | Format | Size | viewBox | Ratio | Use Case |
+|----|--------|------|---------|-------|----------|
+| `ppt169` | PPT 16:9 | `1280x720` | `0 0 1280 720` | 16:9 | Business presentations, meetings |
+| `ppt43` | PPT 4:3 | `1024x768` | `0 0 1024 768` | 4:3 | Traditional projectors, academic talks |
+| `xiaohongshu` | Xiaohongshu (RED) | `1242x1660` | `0 0 1242 1660` | 3:4 | Image-text sharing, knowledge posts |
+| `moments` | WeChat Moments / IG | `1080x1080` | `0 0 1080 1080` | 1:1 | Square posters, brand showcases |
+| `story` | Story / TikTok | `1080x1920` | `0 0 1080 1920` | 9:16 | Vertical stories, short video covers |
+| `wechat` | WeChat Article Header | `900x383` | `0 0 900 383` | 2.35:1 | WeChat article cover images |
+| `banner` | Landscape Banner | `1920x1080` | `0 0 1920 1080` | 16:9 | Web banners, digital screens |
+| `a4` | A4 Print | `1240x1754` | `0 0 1240 1754` | 1:sqrt(2) | Print posters, flyers |
+
+`ppt169` is the canonical PPT wide-screen canvas in this repo: `1280x720`, not any arbitrary 16:9 size. Same-ratio canvases such as `banner` (`1920x1080`) must be treated as different coordinate systems.
## Format Selection Decision Tree
diff --git a/skills/ppt-master/references/template-designer.md b/skills/ppt-master/references/template-designer.md
index dec0df80..37b88825 100644
--- a/skills/ppt-master/references/template-designer.md
+++ b/skills/ppt-master/references/template-designer.md
@@ -130,6 +130,13 @@ summary: <one-line tone & use case>
keywords: [tag1, tag2, tag3]
primary_color: "#......"
canvas_format: ppt169
+canvas_width: 1280
+canvas_height: 720
+canvas_viewbox: "0 0 1280 720"
+# Required when a PPTX/SVG source canvas is known; keep equal to canvas_* unless explicitly normalized.
+source_canvas_width: 1280
+source_canvas_height: 720
+source_viewbox: "0 0 1280 720"
replication_mode: standard | fidelity | mirror
# Optional — only when this template overrides canonical placeholder vocabulary.
# Omit entirely for `mirror` mode (mirror has no placeholders).
@@ -187,7 +194,8 @@ When rewriting an existing template that contains the omitted sections, delete t
### 2. Inherit Design Specification
Templates must strictly follow the finalized template brief and the generated `design_spec.md`:
-- **Canvas dimensions**: viewBox matches the design spec
+- **Canvas dimensions**: `canvas_format` is not enough; root SVG `width` / `height` / `viewBox` match `canvas_width`, `canvas_height`, and `canvas_viewbox` in the design spec
+- **Source canvas**: when a PPTX/SVG reference is used, record `source_canvas_width`, `source_canvas_height`, and `source_viewbox`. If the output canvas differs from the source, normalize all geometry, typography, line heights, strokes, and image crop coordinates explicitly instead of relying on the shared aspect ratio.
- **Color scheme**: Uses primary, secondary, and accent colors from the spec
- **Font plan**: Uses the per-role font families declared in the spec
- **Layout principles**: Margins and spacing conform to the spec
diff --git a/skills/ppt-master/scripts/confirm_ui/static/app.js b/skills/ppt-master/scripts/confirm_ui/static/app.js
index 00829f9e..d09da2f8 100644
--- a/skills/ppt-master/scripts/confirm_ui/static/app.js
+++ b/skills/ppt-master/scripts/confirm_ui/static/app.js
@@ -326,7 +326,7 @@
// Render an enumerable field: ALL options from the catalog, recommended one
// badged, current selection from STATE, plus a trailing Custom box.
- // `list` is either a flat array of {id,label,desc,dim} or a grouped array
+ // `list` is either a flat array of {id,label,desc,dim,viewbox} or a grouped array
// of {group, items:[...]}.
function enumField(parent, list, recommendedId, getVal, setVal, opts2) {
list = list || [];
@@ -364,12 +364,16 @@
function deselect() { allChips.forEach(function (c) { c.classList.remove("selected"); }); }
function makeChip(o) {
var label = optionLabel(o);
- if (o.dim) label += " · " + o.dim;
var desc = optionDesc(o);
- if (desc) label += (LANG === "zh" ? ":" : " — ") + desc;
var spec = specById[o.id];
- if (spec && spec.note) label += " · " + spec.note;
var chip = el("div", "chip");
+ if (o.viewbox) {
+ label = o.id + (o.dim ? " · " + o.dim : "");
+ } else {
+ if (o.dim) label += " · " + o.dim;
+ if (desc) label += (LANG === "zh" ? ":" : " — ") + desc;
+ if (spec && spec.note) label += " · " + spec.note;
+ }
chip.appendChild(el("span", "chip-text", label));
if (spec) {
// spectrum pick: badge shows its temperament tag, not the generic ★
diff --git a/skills/ppt-master/scripts/confirm_ui/static/catalogs.json b/skills/ppt-master/scripts/confirm_ui/static/catalogs.json
index 73686c48..8560c83f 100644
--- a/skills/ppt-master/scripts/confirm_ui/static/catalogs.json
+++ b/skills/ppt-master/scripts/confirm_ui/static/catalogs.json
@@ -7,6 +7,7 @@
"label_zh": "宽屏演示 16:9",
"label_en": "PPT 16:9",
"dim": "1280×720",
+ "viewbox": "0 0 1280 720",
"use_zh": "现代投影 / 在线演示",
"use_en": "Modern projectors / online"
},
@@ -16,6 +17,7 @@
"label_zh": "传统演示 4:3",
"label_en": "PPT 4:3",
"dim": "1024×768",
+ "viewbox": "0 0 1024 768",
"use_zh": "传统投影",
"use_en": "Traditional projectors"
},
@@ -25,6 +27,7 @@
"label_zh": "微信公众号头图",
"label_en": "WeChat Article Header",
"dim": "900×383",
+ "viewbox": "0 0 900 383",
"use_zh": "微信公众号文章封面",
"use_en": "WeChat article cover"
},
@@ -34,6 +37,7 @@
"label_zh": "小红书图文",
"label_en": "Xiaohongshu (RED)",
"dim": "1242×1660",
+ "viewbox": "0 0 1242 1660",
"use_zh": "知识分享 / 种草封面",
"use_en": "Knowledge sharing"
},
@@ -43,6 +47,7 @@
"label_zh": "社交方图",
"label_en": "WeChat Moments / IG",
"dim": "1080×1080",
+ "viewbox": "0 0 1080 1080",
"use_zh": "社交媒体方图",
"use_en": "Social square"
},
@@ -52,6 +57,7 @@
"label_zh": "竖屏短视频",
"label_en": "Story / TikTok",
"dim": "1080×1920",
+ "viewbox": "0 0 1080 1920",
"use_zh": "短视频 / 竖屏封面",
"use_en": "Short-video / story cover"
},
@@ -61,6 +67,7 @@
"label_zh": "横幅",
"label_en": "Landscape Banner",
"dim": "1920×1080",
+ "viewbox": "0 0 1920 1080",
"use_zh": "网页横幅 / 大屏",
"use_en": "Web banner / big screen"
},
@@ -70,6 +77,7 @@
"label_zh": "A4 打印",
"label_en": "A4 Print",
"dim": "1240×1754",
+ "viewbox": "0 0 1240 1754",
"use_zh": "打印 / 文档导出",
"use_en": "Print / PDF export"
}
diff --git a/skills/ppt-master/workflows/create-template.md b/skills/ppt-master/workflows/create-template.md
index 31ed5bed..a84dfb19 100644
--- a/skills/ppt-master/workflows/create-template.md
+++ b/skills/ppt-master/workflows/create-template.md
@@ -126,7 +126,7 @@ Before composing Step 2, extract the template's reusable norms from the previous
| Norm area | Extract from | Record as |
|---|---|---|
-| Canvas / page geometry | `manifest.json` slide size, SVG `viewBox` | `[fact]` canvas format and aspect ratio |
+| Canvas / page geometry | `manifest.json` slide size, SVG `width` / `height` / `viewBox` | `[fact]` canvas format, pixel dimensions, source `viewBox`, and aspect ratio |
| Identity system | theme colors, font usage, logo / emblem assets, recurring backgrounds | `[fact]` when imported; `[suggested]` only for visual estimates |
| Layout grammar | masters / layouts, repeated chrome, margins, columns, card grids, section dividers | Template-specific rules, not generic spacing boilerplate |
| Image system | image crops, masks, full-bleed zones, hero-image placement, mosaic rules, caption / overlay treatment | Template-specific image-placement rules with source examples |
@@ -202,7 +202,7 @@ Compose a single message that surfaces every Required brief item to the user, **
| Template kind | Recommended localized label with English ID, plus both options and the rule for choosing |
| Category | Recommended localized category with English ID, plus `brand` / `general` / `scenario` / `government` / `special` with localized explanations |
| Theme mode | Recommended localized mode with English ID, plus available modes such as `light` / `dark` / `mixed` with localized explanations |
-| Canvas format | Recommended canvas, plus other supported formats from [`canvas-formats.md`](../references/canvas-formats.md) that fit the source aspect ratio or user intent |
+| Canvas format | Recommended canvas, plus other supported formats from [`canvas-formats.md`](../references/canvas-formats.md) that fit the source aspect ratio or user intent. Always show the concrete pixel size and `viewBox`; do not treat two same-ratio formats such as `ppt169` (`1280x720`) and `banner` (`1920x1080`) as interchangeable. |
| Replication mode | Recommended localized mode with English ID, plus all modes available for the current input type; list forbidden modes for type C / D as unavailable with localized reasons |
| Visual fidelity for fixed pages | Recommended localized choice with English ID, plus both `literal` / `adapted` options when applicable |
| Asset bundling | Recommended included assets, plus excluded candidate assets with a one-line reason when reference assets exist |
@@ -217,7 +217,7 @@ Items to surface:
| Applicable scenarios | Yes | `[suggested]` from analysis; user confirms |
| Tone summary | Yes | `[suggested]` from analysis (e.g. `Modern, restrained, data-driven`) |
| Theme mode | Yes | A: `[fact]` from `manifest.json` background colors. B: `[fact]` from SVG `fill`. C: `[suggested]` from visual estimate. D: `[decision]` |
-| Canvas format | Yes | A/B: `[fact]` from slide size or SVG `viewBox`. C: `[suggested]` from image aspect ratio. D: `[decision]`, default `ppt169` |
+| Canvas format and dimensions | Yes | A/B: `[fact]` from slide size or SVG `width` / `height` / `viewBox`; show `canvas_format`, `canvas_width`, `canvas_height`, `canvas_viewbox`, and `source_viewbox`. C: `[suggested]` from image aspect ratio. D: `[decision]`, default `ppt169` (`1280x720`, `0 0 1280 720`) |
| Replication mode | Yes | `[decision]` — `standard` always available; `fidelity` and `mirror` available for type A (canonical, manifest-anchored) and type B (AI visual clustering / direct 1:1 copy — see Step 1 caveats); reject `fidelity` / `mirror` upfront for type C / D |
| Visual fidelity for fixed pages | Yes for `standard` / `fidelity` when reference exists; **N/A for `mirror`** (mirror is implicitly literal) | `[decision]` — `literal` (preserve original geometry / decoration / sprite crops as-is; for cover / chapter / ending especially) or `adapted` (use the reference for tone/structure but allow design evolution). Different page types may take different settings |
| Basic template norms | Yes when reference exists | `[fact]` / `[suggested]` — layout grammar, image system, density rhythm, page roster semantics, and asset policy extracted in Step 1 |
@@ -235,7 +235,7 @@ For type A, also include in this message:
The user replies with corrections, additions, or "all good".
-> **Persist the brief into `design_spec.md`**. When the Template_Designer writes `design_spec.md` in Step 4, declare a YAML frontmatter block at the top with the confirmed brief (`template_id`, `category`, `summary`, `keywords`, `primary_color`, `canvas_format`, `replication_mode`, etc.). `register_template.py` reads this in Step 6, so the brief flows directly into the index without the AI re-deriving it from prose. See Step 6 for the recommended frontmatter shape.
+> **Persist the brief into `design_spec.md`**. When the Template_Designer writes `design_spec.md` in Step 4, declare a YAML frontmatter block at the top with the confirmed brief (`template_id`, `category`, `summary`, `keywords`, `primary_color`, `canvas_format`, `canvas_width`, `canvas_height`, `canvas_viewbox`, `source_viewbox`, `replication_mode`, etc.). `register_template.py` reads this in Step 6, so the brief flows directly into the index without the AI re-deriving it from prose. See Step 6 for the recommended frontmatter shape.
---
@@ -350,7 +350,8 @@ python3 skills/ppt-master/scripts/svg_quality_checker.py "skills/ppt-master/temp
- [ ] Every page declared in `design_spec.md §V Page Roster` exists as an SVG file in the template directory (and vice versa — no orphan files)
- [ ] Variant filenames follow the letter-suffix convention (e.g. `03a_content_two_col.svg`); variants typically reuse the parent type's placeholder set unless the spec frontmatter declares otherwise
- [ ] If TOC exists, placeholder pattern uses the canonical indexed form
-- [ ] SVG viewBox matches the chosen canvas format (for `ppt169`: `0 0 1280 720`)
+- [ ] `design_spec.md` frontmatter declares `canvas_format`, `canvas_width`, `canvas_height`, and `canvas_viewbox`; PPTX/SVG-backed templates also declare `source_canvas_width`, `source_canvas_height`, and `source_viewbox`
+- [ ] SVG `width` / `height` / `viewBox` match the declared canvas dimensions, not just the aspect ratio (for `ppt169`: `1280x720`, `0 0 1280 720`; for `banner`: `1920x1080`, `0 0 1920 1080`)
- [ ] Placeholder names follow the canonical convention where applicable; templates with intentionally different vocabularies (e.g. `{{KEY_MESSAGE}}` instead of `{{PAGE_TITLE}}`) should declare a `placeholders:` frontmatter block to silence advisory warnings
- [ ] Asset files referenced by SVGs actually exist in the template package
- [ ] If any SVG references an extracted vector `data-icon`, the corresponding SVG asset exists directly under the package's `icons/` directory; do not add a separate illustration embedding script
@@ -394,6 +395,12 @@ The index file is a **discovery index** — it lets the AI answer "what template
> kind: deck
> summary: ...
> canvas_format: ppt169
+> canvas_width: 1280
+> canvas_height: 720
+> canvas_viewbox: "0 0 1280 720"
+> source_canvas_width: 1280
+> source_canvas_height: 720
+> source_viewbox: "0 0 1280 720"
> page_count: 5
> primary_color: "#005587"
> ---
@@ -404,6 +411,12 @@ The index file is a **discovery index** — it lets the AI answer "what template
> kind: layout
> summary: ...
> canvas_format: ppt169
+> canvas_width: 1280
+> canvas_height: 720
+> canvas_viewbox: "0 0 1280 720"
+> source_canvas_width: 1280
+> source_canvas_height: 720
+> source_viewbox: "0 0 1280 720"
> page_count: 5
> page_types: [cover, toc, chapter, content, ending]
> ---