# S-24 Slide Prompt V2

## Status

- Slide/version: `S-24 / s24-v2`.
- Status: prompt ready for host-native generation.
- Worker: Agent S-24, session-11 retry.
- Overwrite allowed: no.
- Review: disabled; do not create `s24-v2-review.md`.

## Based On Brief

- Brief: `projects/2026-06-23-在路上/slides/s24/s24-v2-brief.md`.
- Slide function: turn `也曾身陷无名旷野 / 被世界吞没` into spatial relationship.
- First read: huge unnamed wilderness before the tiny traveler.

## Reference Inputs

- Image 1: common style calibration board, `projects/2026-06-23-在路上/visual-system/style/style-anchor-board-v1.png`; role: style calibration, low-saturation documentary road-film grammar.
- Image 2: A landscape-led blueprint, `projects/2026-06-23-在路上/intake/imgs/A.风景主导型蓝图.png`; role: primary landscape-led composition, world-before-person, rough road and wilderness scale.
- Image 3: distant walking-back silhouette reference, `projects/2026-06-23-在路上/visual-system/characters/character-slice-walk-back-v2-from-anchor.png`; role: silhouette continuity guard only; do not enlarge for identity readability.

## Advisor / Host-Native Record

- Skill: `gpt-image-2`.
- Mode check command: `node C:\Users\wangq\.agents\skills\gpt-image-2\scripts\check-mode.js --json`.
- Mode check result: `mode=B-or-C`, `recommendation=host-or-advisor`, `garden_mode_enabled=false`, `has_api_key=false`, `model=gpt-image-2`.
- Effective mode: Mode B / host-native, because the host image generation tool is available.
- Garden/API expected: no.
- Skill references read:
  - `C:\Users\wangq\.agents\skills\gpt-image-2\SKILL.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`
- Project source files read:
  - `AGENTS.md`
  - `docs/workflows/slide-task-orchestration.md`
  - `docs/workflows/slide-task-templates.md`
  - `projects/2026-06-23-在路上/tasks/2026-06-24-character-consistency/session-11/packets/s24-v2-slide-agent-packet.md`
  - `projects/2026-06-23-在路上/slides/s24/s24-v2-brief.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/intake/阶段五-镜头清单骨架.md`

## Final Prompt

Create one horizontal 16:9 photorealistic cinematic documentary road-film still.

Use Image 1 as the overall style calibration: real documentary texture, restrained poetic realism, low saturation, cold gray, earth brown, dark khaki, coal black, wet blue-gray, no heroic or commercial polish. Use Image 2 as the primary composition and landscape reference: a landscape-led road scene where the world appears before the person, with rough mud, ruts, frost, stones, sparse terrain, a subdued sky, and a long worn road swallowed by wilderness. Use Image 3 only as a tiny walking-away silhouette continuity guard: back view, long dark coat, backpack, walking posture, but the traveler must remain nearly a point.

Scene: an extreme wide shot of a huge unnamed wilderness after cold rain or frost, rough muddy earth stretching across most of the frame, sparse low vegetation, stones, shallow puddles, dull sky pressure, and a long road or track that loses itself into distance. The landscape must dominate so strongly that the person feels trapped inside scale, distance, emptiness, and terrain.

Subject scale and placement: place one lone traveler walking away along the muddy track, around 1/40 to 1/25 of the frame, almost a dot. The traveler should be readable only as a small dark back silhouette with a backpack and long coat, not as a portrait. Do not enlarge the person for identity readability. No face, no facial drama, no centered hero stance.

Composition: environment first, road/terrain second, traveler last. Use deep distance, negative space, low horizon or wide sky/ground pressure, and a broad landscape-led frame. Oppression must come from spatial relationship: the world is larger than the person, the road is longer than the person, and the traveler is only a moving coordinate.

Lighting and texture: cold muted dawn or overcast pre-dawn, subdued natural light, muddy surface detail, wet ruts, frost edges, rough stones, restrained haze. Warmth, if any, should be extremely faint and only an atmospheric signal, not a sunrise spectacle.

Avoid: travel advertising, scenic tourism, inspirational road poster, beautiful postcard vista, broad sunrise or sunset, fantasy landscape, magic glow, heroic centered backshot, fashion photography, portrait MV, dramatic face, enlarged character, clean road, clean commercial city, camping lifestyle, animals, vehicles, text, subtitles, logos, labels, UI, watermark.

## Output Intent

- Generate exactly one image.
- Target project image: `projects/2026-06-23-在路上/slides/s24/s24-v2-image.png`.
- Required visual result: extreme wide landscape-led wilderness; tiny traveler nearly swallowed by the world.