docs(image-text): drop stale "CJK fails / English-only" capability claims

In-image AI text guidance gated the embedded-vs-SVG decision on language
(CJK fails) and length (long text garbles). Both are outdated, backend-
specific capability assertions — current backends (gpt-image-2 etc.) render
CJK and long strings fine. Reframe around the one durable, backend-independent
fact: Layer 1 text is rasterized and cannot be edited, so the discriminator is
editability/exactness, not language or length. Verify the rendered output
instead of predicting failure.

Swept: image-generator.md §5.3 + Primitive C, image-layout-patterns #64/#65,
design_spec_reference.md text_policy, and the chalkboard/sketch-notes/ink-notes
+ comparison/framework/infographic/flowchart fewshot snippets. CHART_STYLE_GUIDE
§8.0 (chart-template placeholder English convention) is unrelated, left as-is.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Author: hugohe3

skills/ppt-master/references/image-generator.md

@@ -179,7 +179,7 @@ Use for: founder profile, speaker bio, testimonial page, executive intro. Pair w
 
 > The image's central content is one large text element — a short headline, big number, or single word — rendered as art, occupying 40-60% of canvas height. Minimal supporting visual (small icon, geometric anchor, accent line) at <25% weight. At least 20% padding around the text.
 
-Use with `text_policy: embedded`. Must obey the §5.3 text-accuracy rules (English short-text and simple numbers reliable; CJK and long sentences fail — for those, switch to Primitive D + SVG overlay).
+Use with `text_policy: embedded`. Must obey the §5.3 rule — text that is part of the artwork and stable can be embedded; copy that must stay exact or editable goes to SVG overlay (switch to Primitive D). Verify the rendered text in the output.
 
 **Primitive D — atmospheric backdrop (no subject)**
 
@@ -303,20 +303,20 @@ The table below is **a reference for the one case where you want the in-image le
 
 **When to use the table**: a designed title (cover main title, chapter heading) on a deck whose visual identity is grounded in the SVG body typography, and where a surprise font choice would feel out of place.
 
-**In-image text accuracy — what image models can and cannot render reliably**
+**In-image text vs SVG text — decide by editability, not by model capability**
 
-Image models have **variable accuracy on text rendering**. Pick the in-image text content with this in mind, especially for §4.1 Primitive C (typographic hero) where the word *is* the image:
+Layer 1 text is rasterized into the artwork — once generated it cannot be edited, corrected, searched, restyled, or reflowed. That is the durable reason to choose where text lives, independent of any backend's rendering ability or the script / length involved:
 
-| Content | Reliability |
+| Text | Layer |
 |---|---|
-| English short text (1-5 characters, one short word) | High — most modern models render correctly most of the time |
-| English longer text (full sentences) | Low — typos, glyph errors, dropped letters |
-| Simple numbers / symbols (`100`, `5x`, `$50M`, `∞`, `?`) | Variable — verify after generation, often fail at large display scale |
-| Chinese / Japanese / Korean characters | Very low — most models fail consistently |
+| Part of the artwork and stable — decorative lettering, designed title, hand-lettered keyword, figure-internal identifiers (axis labels, panel letters, units) | Layer 1 (image) OK |
+| Page chrome, body copy, captions, data values — anything that must stay exact, searchable, or may be reworded | Layer 2 (SVG) |
 
-**Prefer in-image**: a short English word (1-2 words max), a simple number, a single symbol or letter.
+Generation is non-deterministic on every backend — **always verify the rendered text in the output** and regenerate if a glyph came out wrong. Do not assume a script or a length will fail; check the actual result. For §4.1 Primitive C (typographic hero) the word *is* the image, so verifying the output matters most there.
 
-**Push to SVG overlay instead**: long quotes / sentences, CJK headlines, complex composite text. When the desired headline is long or CJK, switch to **Primitive D (atmospheric backdrop)** and overlay the headline as SVG text.
+**Prefer in-image**: text that is genuinely part of the artwork and will not be edited — a designed word, a stat lettering, a figure-internal label.
+
+**Push to SVG overlay instead**: page chrome, captions, data values, or any copy that must stay exact or editable. When the headline must remain editable, switch to **Primitive D (atmospheric backdrop)** and overlay it as SVG text.
 
 ### 5.4 No brand names or trademarks in the subject
 

skills/ppt-master/references/image-layout-patterns.md

@@ -172,9 +172,9 @@ Stack any of these freely on top of a Primary structure. Multiple Modifiers per
 
 63. **Transparent PNG sticker / cutout** — an RGBA PNG (with alpha channel) placed via standard `<image>` — no `clipPath` required, the transparency lives in the file itself. Useful for subjects that should not appear inside a rectangular frame (people cutouts, product shots, decorative motifs floating over backgrounds). Producing transparent PNGs is **not** a standard ppt-master pipeline step — three paths: (a) AI backend that supports transparent output natively, (b) generate a chroma-key (solid green background) image then strip the green with a separate tool, (c) user-supplied transparent asset. SVG-side usage is trivial; asset preparation is the work.
 
-64. **Image with embedded text rendered by the AI** — text becomes part of the artwork: decorative lettering, designed title, hand-lettered keyword. Prompt with explicit text content. Reliable only for **short English**. Chinese characters, long phrases, and digits regularly come back malformed and cannot be edited. Anything that must be correct or editable goes in the SVG `<text>` layer.
+64. **Image with embedded text rendered by the AI** — text becomes part of the artwork: decorative lettering, designed title, hand-lettered keyword. Prompt with explicit text content. Use for text that is part of the artwork and will not change; verify the rendered text in the output and regenerate if a glyph came out wrong. Anything that must be correct or editable goes in the SVG `<text>` layer (#65).
 
-65. **Image with NO text — labels added as native SVG** — generate the image with explicit "no text, no letters, no numbers, no signs" instruction (`text_policy: none`), then place all labels as `<text>` overlays. The right call when labels are CJK, will be reworded, or carry data that must stay editable — pair with `#64` when stable visual identifiers (axis labels, subplot letters, unit symbols) belong inside the image instead.
+65. **Image with NO text — labels added as native SVG** — generate the image with explicit "no text, no letters, no numbers, no signs" instruction (`text_policy: none`), then place all labels as `<text>` overlays. The right call when labels will be reworded, must stay exact, or carry data that must stay editable — pair with `#64` when stable visual identifiers (axis labels, subplot letters, unit symbols) belong inside the image instead.
 
 66. **Image fading into the solid background** — soften the image's edge into the deck's background color via a `<linearGradient>` overlay whose end-stop matches the background hex exactly. The image's rectangular boundary disappears, producing seamless integration.
 

skills/ppt-master/references/image-renderings/chalkboard.md

@@ -32,4 +32,4 @@ chalkboard has a strong **fixed-palette tendency**: dark slate background + whit
 
 **Snippet A — half-page lesson visual, text_policy: embedded**
 
-> Classroom chalkboard aesthetic. Dark slate green background `#1F3A2E` with visible chalk-dust texture at 20% opacity. Three rounded boxes drawn in soft white chalk strokes arranged in a row, each containing a simple chalk-drawn icon (a tree, a lightbulb, a paper plane). Above the boxes, a hand-lettered chalk title in confident white lettering reads "GROWTH" in English block caps. Hand-drawn chalk arrows connect the boxes left-to-right. Small chalk star and dot decorations sprinkled sparingly. One soft pastel-yellow chalk underline below the title, drawn with deliberate dry-medium quality. Composed as a 600×500 half-page block with 14% inner padding. NO Chinese characters — chalk text in short English only. Color values are rendering guidance only.
+> Classroom chalkboard aesthetic. Dark slate green background `#1F3A2E` with visible chalk-dust texture at 20% opacity. Three rounded boxes drawn in soft white chalk strokes arranged in a row, each containing a simple chalk-drawn icon (a tree, a lightbulb, a paper plane). Above the boxes, a hand-lettered chalk title in confident white lettering reads "GROWTH" in English block caps. Hand-drawn chalk arrows connect the boxes left-to-right. Small chalk star and dot decorations sprinkled sparingly. One soft pastel-yellow chalk underline below the title, drawn with deliberate dry-medium quality. Composed as a 600×500 half-page block with 14% inner padding. Chalk text kept to short keywords. Color values are rendering guidance only.

skills/ppt-master/references/image-renderings/ink-notes.md

@@ -34,4 +34,4 @@ This makes ink-notes **the rendering most likely to deviate from the deck's HEX*
 
 **Snippet A — Before/After methodology (comparison type), text_policy: embedded**
 
-> Professional hand-drawn visual-note style on pure white background. Composition is a Before/After split — vertical hand-drawn divider down the center. Both sides use black ink line work with slight wobble. Left side ("Before") shows a simple stick-figure character with a frustrated posture, a speech bubble with hand-lettered "OLD WAY" in English block caps, and a small list of three hand-drawn dashes with brief 1-2 word annotations (e.g. "manual", "slow", "fragile"). Right side ("After") shows a confident stick-figure character with a clean checkmark above, hand-lettered "NEW WAY" in English block caps, and three checkbox-style annotations (e.g. "automated", "fast", "reliable"). A curved hand-drawn "mindset shift" arrow bridges left to right with a small hand-lettered label "the shift". Sparse semantic color: coral red `#E8655A` (the deck's accent) appears only on the left side's pain points; muted teal `#5FA8A8` appears only on the right side's positives. Total color accent area under 10% of canvas. NO Chinese characters — all hand-lettered text is short English keywords. Composed as a 1200×500 hero banner with 14% inner padding.
+> Professional hand-drawn visual-note style on pure white background. Composition is a Before/After split — vertical hand-drawn divider down the center. Both sides use black ink line work with slight wobble. Left side ("Before") shows a simple stick-figure character with a frustrated posture, a speech bubble with hand-lettered "OLD WAY" in English block caps, and a small list of three hand-drawn dashes with brief 1-2 word annotations (e.g. "manual", "slow", "fragile"). Right side ("After") shows a confident stick-figure character with a clean checkmark above, hand-lettered "NEW WAY" in English block caps, and three checkbox-style annotations (e.g. "automated", "fast", "reliable"). A curved hand-drawn "mindset shift" arrow bridges left to right with a small hand-lettered label "the shift". Sparse semantic color: coral red `#E8655A` (the deck's accent) appears only on the left side's pain points; muted teal `#5FA8A8` appears only on the right side's positives. Total color accent area under 10% of canvas. All hand-lettered text is short keywords. Composed as a 1200×500 hero banner with 14% inner padding.

skills/ppt-master/references/image-renderings/sketch-notes.md

@@ -33,4 +33,4 @@ sketch-notes has a strong **built-in palette tendency** toward warm cream + blac
 
 **Snippet A — half-page educational concept, text_policy: embedded**
 
-> Warm hand-drawn sketchnote on warm cream paper background. Black ink lines with slight wobble define three rounded rectangle info boxes arranged in a soft triangle layout. Each box is filled with a soft pastel block color: top box in light blue (a pastel tint of the deck's primary `#1E3A5F`), bottom-left box in mint, bottom-right box in lavender. Color fills don't completely reach the outlines (slight hand-painted overshoot). Hand-drawn wavy arrows connect the boxes — each arrow with a small inline hand-lettered English keyword like "leads to", "becomes", "supports" (no Chinese characters; ≤2 words per arrow). Each box contains one simple hand-drawn cartoon icon — a lightbulb, a plant, a gear — in black ink. Small doodle decorations (a few stars, dots, sparkles) sparingly around the composition. Composed as a 600×600 half-page block with 14% inner padding. Generous white space. Color values are rendering guidance only — do not display HEX codes or color names as text.
+> Warm hand-drawn sketchnote on warm cream paper background. Black ink lines with slight wobble define three rounded rectangle info boxes arranged in a soft triangle layout. Each box is filled with a soft pastel block color: top box in light blue (a pastel tint of the deck's primary `#1E3A5F`), bottom-left box in mint, bottom-right box in lavender. Color fills don't completely reach the outlines (slight hand-painted overshoot). Hand-drawn wavy arrows connect the boxes — each arrow with a small inline hand-lettered keyword like "leads to", "becomes", "supports" (≤2 words per arrow). Each box contains one simple hand-drawn cartoon icon — a lightbulb, a plant, a gear — in black ink. Small doodle decorations (a few stars, dots, sparkles) sparingly around the composition. Composed as a 600×600 half-page block with 14% inner padding. Generous white space. Color values are rendering guidance only — do not display HEX codes or color names as text.

skills/ppt-master/references/image-type-templates/comparison.md

@@ -45,7 +45,7 @@ Each side contains a visual anchor (figure, icon, scene) but no text labels. SVG
 
 ### `text_policy: embedded`
 
-Each side may contain short hand-lettered labels: "BEFORE" / "AFTER" at the top, plus 2-3 short English bullet words inside (e.g. "slow / manual / fragile" vs "fast / automated / reliable").
+Each side may contain short hand-lettered labels: "BEFORE" / "AFTER" at the top, plus 2-3 short bullet words inside (e.g. "slow / manual / fragile" vs "fast / automated / reliable").
 
 > ink-notes + comparison + embedded is a classic combination — the "manifesto Before/After" page.
 
@@ -55,4 +55,4 @@ Each side may contain short hand-lettered labels: "BEFORE" / "AFTER" at the top,
 
 **Snippet A — ink-notes + mono-ink, Before/After horizontal split, text_policy: embedded, 1200×500**
 
-> Professional hand-drawn visual-note style on pure white background `#FFFFFF`. The composition is a Before/After horizontal split with a clean vertical hand-drawn divider down the center. Both sides use confident black ink line work with slight wobble. Left side ("Before"): a simplified stick-figure character with frustrated posture, a speech bubble in hand-lettered English caps reading "OLD WAY", and three small hand-drawn dashes with brief 1-word annotations ("manual", "slow", "fragile"). Right side ("After"): a confident stick-figure character with a checkmark above, hand-lettered "NEW WAY" caps label, and three checkbox-style annotations ("automated", "fast", "reliable"). A curved "shift" arrow bridges left to right at mid-height. Sparse semantic accent: coral red `#E8655A` only on left-side pain markers (about 4% area); muted teal `#5FA8A8` only on right-side positive markers (about 3% area). Total color accent under 8%. Composed as a 1200×500 hero banner with 14% inner padding. English-only hand-lettered text. NO Chinese characters. Color values are rendering guidance only.
+> Professional hand-drawn visual-note style on pure white background `#FFFFFF`. The composition is a Before/After horizontal split with a clean vertical hand-drawn divider down the center. Both sides use confident black ink line work with slight wobble. Left side ("Before"): a simplified stick-figure character with frustrated posture, a speech bubble in hand-lettered English caps reading "OLD WAY", and three small hand-drawn dashes with brief 1-word annotations ("manual", "slow", "fragile"). Right side ("After"): a confident stick-figure character with a checkmark above, hand-lettered "NEW WAY" caps label, and three checkbox-style annotations ("automated", "fast", "reliable"). A curved "shift" arrow bridges left to right at mid-height. Sparse semantic accent: coral red `#E8655A` only on left-side pain markers (about 4% area); muted teal `#5FA8A8` only on right-side positive markers (about 3% area). Total color accent under 8%. Composed as a 1200×500 hero banner with 14% inner padding. Hand-lettered text kept to short keywords. Color values are rendering guidance only.

skills/ppt-master/references/image-type-templates/flowchart.md

@@ -57,7 +57,7 @@ Each stage contains an iconic symbol only. Stage labels added in SVG overlay bel
 
 ### `text_policy: embedded`
 
-Each stage may contain a short English keyword (≤2 words) inside or beside it.
+Each stage may contain a short keyword (≤2 words) inside or beside it.
 
 ---
 

skills/ppt-master/references/image-type-templates/framework.md

@@ -69,7 +69,7 @@ A short keyword (1-2 English words) appears inside or beside each satellite / ce
 
 Sample fragment:
 
-> Each satellite includes a single short hand-lettered keyword in English (≤2 words) — e.g. "data", "process", "growth". Keywords are part of the artwork, not labels. No long sentences, no numbers, no Chinese characters (most models render CJK characters incorrectly).
+> Each satellite includes a single short hand-lettered keyword (≤2 words) — e.g. "data", "process", "growth". Keywords are part of the artwork, not labels. No long sentences, no numbers; keep each keyword short.
 
 ---
 

skills/ppt-master/references/image-type-templates/infographic.md

@@ -46,11 +46,11 @@ Sample fragment:
 
 ### `text_policy: embedded`
 
-Each zone contains an icon + one short English keyword (≤2 words).
+Each zone contains an icon + one short keyword (≤2 words).
 
 Sample fragment:
 
-> Each zone contains one iconic symbol and one short hand-lettered English keyword (1-2 words, e.g. "Speed", "Reach", "Trust"). Keywords are part of the artwork. No long sentences, no numbers, no Chinese characters.
+> Each zone contains one iconic symbol and one short hand-lettered keyword (1-2 words, e.g. "Speed", "Reach", "Trust"). Keywords are part of the artwork. No long sentences, no numbers; keep each keyword short.
 
 ---
 

skills/ppt-master/templates/design_spec_reference.md

@@ -276,7 +276,7 @@ Catalog read: 71 templates
 **text_policy** (`ai` rows only; AI judges per row, no global default bias):
 
 - `none` — image carries no text; SVG overlays all labels
-- `embedded` — image contains in-artwork text: decorative lettering, a designed title, hand-lettered keywords, or stable visual identifiers (axis labels, subplot letters, unit symbols). Body copy / data points / long quotes never go inside the image regardless. English text renders most reliably; CJK characters fail in most models
+- `embedded` — image contains in-artwork text: decorative lettering, a designed title, hand-lettered keywords, or stable visual identifiers (axis labels, subplot letters, unit symbols). Body copy / data points / long quotes never go inside the image regardless — they must stay editable. Embedded text is frozen into the raster; verify the rendered text in the output
 
 **page_role** (`ai` rows only; leave blank for default):