115 lines
5.6 KiB
Markdown
115 lines
5.6 KiB
Markdown
# S-11 V2 Prompt
|
|
|
|
## Status
|
|
|
|
- Slide: `S-11`
|
|
- Version: `s11-v2`
|
|
- Running mode: `prompt and generate`
|
|
- Worker: `Agent S-11`
|
|
- Review mode: none
|
|
- Overwrite allowed: no
|
|
|
|
## GPT-Image-2 Skill Mode
|
|
|
|
Mode check command:
|
|
|
|
```powershell
|
|
node C:\Users\wangq\.agents\skills\gpt-image-2\scripts\check-mode.js --json
|
|
```
|
|
|
|
Mode check 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 Advisor / prompt-engineering workflow.
|
|
- Garden/API generation expected: no.
|
|
- Use host built-in `image_gen` after this prompt/spec is written.
|
|
|
|
## Skill Template / Reference Basis
|
|
|
|
Skill files read:
|
|
|
|
- `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\scenes-and-illustrations\concept-scene.md`
|
|
- `C:\Users\wangq\.agents\skills\gpt-image-2\references\scenes-and-illustrations\minimalist-mood-scene.md`
|
|
|
|
Template adaptation:
|
|
|
|
- `concept-scene.md` informs foreground / midground / background layering and wide cinematic scene construction.
|
|
- `minimalist-mood-scene.md` informs restraint, negative space, small subject scale, and mood discipline.
|
|
- Final target is a photorealistic cinematic documentary still, not concept art, illustration, poster, storyboard sheet, or text graphic.
|
|
|
|
## Project Sources Read
|
|
|
|
- `AGENTS.md`
|
|
- `projects/2026-06-23-在路上/slides/s11/s11-v2-brief.md`
|
|
- `projects/2026-06-23-在路上/tasks/2026-06-24-character-consistency/session-07/packets/s11-v2-slide-agent-packet.md`
|
|
- `projects/2026-06-23-在路上/visual-system/style/style-prompt-master.md`
|
|
- `projects/2026-06-23-在路上/visual-system/characters/character-reference-registry.md`
|
|
- `projects/2026-06-23-在路上/slides/slides.md`
|
|
|
|
## Reference Inputs
|
|
|
|
- Style module: `B | 高位观察：系统边缘里的小人物`
|
|
- Image 1: `projects/2026-06-23-在路上/intake/imgs/B.高位观察型蓝图.png`
|
|
- Role: style / composition reference; high-angle observational city-edge pressure; wet ground, cold skyline, large system scale.
|
|
- Image 2: `projects/2026-06-23-在路上/visual-system/characters/character-anchor-v2-walk-side-reference-crop-tight.png`
|
|
- Role: walking-side identity and body crop; coat, backpack, glasses profile, hair/beard silhouette, stride, boots.
|
|
- Image 3: `projects/2026-06-23-在路上/visual-system/characters/character-slice-walk-side-v2-from-anchor.png`
|
|
- Role: full-body walking-side silhouette; coat length, backpack profile, walking posture, boot scale.
|
|
- Generation path note: reference images are visible in the generation context as Image 1, Image 2, and Image 3.
|
|
|
|
## Final Host Generation Prompt
|
|
|
|
Create a 16:9 horizontal photorealistic cinematic documentary road-film still for slide S-11.
|
|
|
|
Use Image 1 as the main style and composition reference: a high observational city-edge frame where the built system is larger than the person. Preserve the feeling of wet gray ground, cold city pressure, broad horizontal scale, distant skyline, hard infrastructure, and a tiny traveler overwhelmed by the environment.
|
|
|
|
Use Image 2 and Image 3 as identity/body continuity references for the visible traveler. The traveler must remain small and secondary, about 1/10 to 1/7 of the frame, read mainly through the dark long coat, worn backpack, side-walking posture, stride, boots, glasses/hair/beard silhouette only if visible, and road-worn body language. Do not turn him into a portrait or fashion subject.
|
|
|
|
Scene:
|
|
|
|
- A cold city edge where glass, concrete, steel, retaining walls, drainage lines, bridge or overpass structure, and wet gray ground form a hard wall rather than shelter.
|
|
- The place should feel like the outside edge of the city: service road, under-bridge margin, flood-control embankment, construction-adjacent concrete, puddles, rust traces, dirty blue-gray reflections, and distant hard skyline.
|
|
- No warm home-like windows. No cozy interior light. No clean commercial CBD.
|
|
|
|
Composition:
|
|
|
|
- Environment first, wall / city edge second, traveler third.
|
|
- Use a high or slightly elevated observational camera, wide 16:9 frame.
|
|
- The traveler is lower or off to the side, not centered, walking along the wet concrete / mud-gray edge beside the hard wall.
|
|
- Strong foreground / midground / background separation: wet ground and drain lines in foreground, cold wall / bridge / retaining structure in midground, distant gray city in background.
|
|
- The frame should feel spacious and oppressive, with the traveler as a moving coordinate inside a much larger system.
|
|
|
|
Light, color, texture:
|
|
|
|
- Cold overcast daylight.
|
|
- Low saturation: steel gray, dirty blue-gray, wet concrete gray, coal black, dark khaki, muted brown rust.
|
|
- Documentary realism, real wet surfaces, puddle reflections, grit, stains, worn concrete, utilitarian infrastructure.
|
|
- No neon, no cyberpunk, no sci-fi, no fantasy, no glossy advertising polish.
|
|
|
|
Emotional first read:
|
|
|
|
Cold glass, concrete, steel, wet ground, a hard city wall, and a tiny traveler who has no place to belong.
|
|
|
|
Avoid:
|
|
|
|
- clean CBD promotion, glass-tower advertisement, cyberpunk, neon, sci-fi city, UI, code, logos, readable text, watermarks
|
|
- heroic centered traveler, portrait MV, fashion travel photo, warm home-like window, tourism poster, motivational sunrise or sunset
|
|
- changing the traveler into a different body type, clean outfit, no backpack, different coat, or front-facing glamour pose
|
|
|
|
Output should look like a real documentary still from a restrained road film, not an illustration, poster, storyboard grid, or concept-art painting.
|