video-workbench/projects/2026-07-01-为什么是我们/visual-system/style/style-prompt-master.md

185 lines
7.1 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# Style Prompt Master
## Status
- Document status: validation-stage style base.
- Project: `2026-07-01-为什么是我们`.
- Purpose: reusable style prompt master for P01/P10 validation and later production.
- Generation status: P01/P10 authorized by user; remaining production waits for validation pass.
## Skill Mode Record
Mode check command:
```powershell
node C:\Users\wangq\.agents\skills\gpt-image-2\scripts\check-mode.js --json
```
Result:
```json
{
"mode": "B-or-C",
"recommendation": "host-or-advisor",
"garden_mode_enabled": false,
"has_api_key": false,
"base_url": "https://api.openai.com/v1",
"model": "gpt-image-2",
"env_flag_value": "(unset)"
}
```
Interpretation:
- Use `gpt-image-2` as prompt-engineering workflow.
- Effective path is Mode B because host `image_gen` is available.
- Do not call Garden generation scripts.
- Save prompts before generation.
## Skill Template / Reference Basis
Skill files read for this validation run:
- `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\prompt-writing.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\dense-explainer-slides.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\visual-report-page.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\slides-and-visual-docs\educational-diagram-slide.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\poster-and-campaigns\campaign-kv.md`
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\typography-and-text-layout\title-safe-poster.md`
- `C:\Users\wangq\.codex\skills\.system\imagegen\SKILL.md`
Template adaptation:
- `dense-explainer-slides.md` informs high-density but readable page logic.
- `visual-report-page.md` informs sober business credibility.
- `educational-diagram-slide.md` informs P10's extraction / validation process model.
- `campaign-kv.md` informs cross-page visual continuity.
- `title-safe-poster.md` informs cover-title safety and large Chinese judgment text.
## Project Sources Actually Used
- `project.md`
- `execution-plan.md`
- `visual-system/visual-system.md`
- `visual-system/motifs/motif-registry.md`
- `intake/阶段3PPT 总纲 v1.md`
- `intake/阶段4通用表达与视觉系统.md`
- `intake/阶段5P1-P22 单页设计骨架.md`
Previous deck references used for style continuity:
- `../2026-06-29-搭上AI这条船/visual-system/visual-system.md`
- `../2026-06-29-搭上AI这条船/visual-system/style/style-prompt-master.md`
- `../2026-06-29-搭上AI这条船/visual-system/motifs/motif-registry.md`
- `../2026-06-29-搭上AI这条船/slides/s01/s01-v1-image.png`
- `../2026-06-29-搭上AI这条船/slides/s10/s10-v1-image.png`
Backup source recorded but not used for generation:
```text
C:\Users\wangq\Documents\Codex\knowledge-vault\work\internal\OPC销售\为什么是我们.md
```
## Global Style Kernel
Every page prompt should inherit this kernel unless validation changes the visual system.
Core visual definition:
> Deep navy-black executive background, AI maritime route-map atmosphere, deepwater business-layer cutaway, radar glow, compass geometry, precision mold / toolbox / evidence motifs, cyan signal light and restrained gold highlights.
Chinese working definition:
> 深蓝黑商务底AI 航线作战图,深水区业务剖面,雷达与罗盘光效,工具箱 / 工厂 / 模具师 / 证据链母题,青色和金色信号高亮。
Default prompt kernel:
```text
Create a 16:9 full-page visual slide image for an internal sales-alignment PowerPoint. Deep navy-black executive business background, AI maritime route-map atmosphere, deepwater business-layer cutaway, radar scan rings, compass lines, subtle coordinate grid, cyan signal light and restrained gold highlights. The page should feel like a sober AI deepwater operation map: clear market judgment, expert capability engineering, and sales confidence. It is a prequel to an internal AI kickoff deck, harder and more analytical than a mobilization cover. Not a customer proposal, not a training manual, not a generic SaaS dashboard.
```
Default text handling:
```text
Include only short, large, high-contrast Chinese slide text: one main title, one core judgment, and a few short labels. Use bold modern Chinese sans-serif. Main title and core judgment must be readable. Avoid long paragraphs.
```
Default negative drift:
```text
Avoid platform-bashing, developer-bashing, customer proposal style, government-report style, dense training manual, cute cartoon ship, pirate or fantasy ship, cyberpunk neon overload, generic tech dashboard UI, humanoid robots, stock-photo meeting people, concrete portraits, decorative ocean scenes with no business logic, unreadable dense text, logos, watermarks.
```
## Page-Family Modules
### Module A: Opening / Problem
Use for P01-P03.
```text
Opening problem page. Use the dark AI route-map base from the previous deck, but move the emotion from "boarding excitement" to "entering AI deepwater". Show route lines, radar targets, deepwater layers, and distant lighthouse signals. The page must feel like a sales confidence prequel: we are here to answer why customers should choose us.
```
### Module B: Market Contrast
Use for P04-P07.
```text
Market contrast page. Show platforms as strong toolbox / factory / infrastructure capacity, while the deeper layer asks what should be built and how expert judgment becomes reliable capability. Keep the tone mature: platform is partner and tool, not enemy.
```
### Module C: Our Positioning
Use for P08-P12.
```text
Positioning page. Show expert agents as role-bound digital workers, and our role as system architect and mold maker for expert capability. Use extraction, encapsulation, validation, calibration, and workflow-entry pipeline. Keep the page weighty and memorable, not decorative.
```
### Module D: Customer Choice
Use for P13-P18.
```text
Customer-choice page. Use route judgment, compass, iceberg, validation route, and calibration loop. The page should prove why customers choose us without claiming we are biggest, cheapest, or万能.
```
### Module E: Sober Judgment
Use for P19-P21.
```text
Sober internal judgment page. Show evidence chain, maturity dashboard, and partially built lighthouses / arsenal. The mood is clear confidence, not overclaiming.
```
### Module F: Sales Arsenal
Use for P22.
```text
Action close page. Show sales arsenal maturity map: positioning layer, proof layer, and operating layer. The page should convert the deck's stance into reusable sales tools.
```
## Per-Page Prompt Requirements
Every `sNN-vN-prompt.md` should include:
- Page and version.
- Page family and active module.
- Core motif from `visual-system/motifs/motif-registry.md`.
- Main title from the accepted skeleton.
- Core judgment from the accepted skeleton.
- Skill mode and template/reference path record.
- Source files actually read.
- Reference image roles if any.
- Fact-check source status.
- Final host-native prompt.
Rules:
- Do not generate from the raw material source.
- Do not change P1-P22 order or main conclusions.
- Do not add long source excerpts.
- Do not generate images before user-authorized page scope.