# S-08 Slide Prompt V3

## Status

- Status: ready for host-native generation.
- Slide: `S-08`.
- Version: `s08-v3`.
- Running mode: prompt and generate.
- Review mode: none; do not write `s08-v3-review.md`.
- Overwrite allowed: no.

## Based On Brief

- Brief: `projects/2026-06-23-在路上/slides/s08/s08-v3-brief.md`.
- Packet: `projects/2026-06-23-在路上/tasks/2026-06-24-character-consistency/session-06/packets/s08-v3-slide-agent-packet.md`.
- Retry note: `s08-v2` is an incomplete transport-failure attempt and is not repaired or reused.

## Reference Inputs

- Style module: character detail with road texture, primarily Module A plus the global style kernel.
- Active style reference: `projects/2026-06-23-在路上/intake/imgs/A.风景主导型蓝图.png`.
- Common calibration board: `projects/2026-06-23-在路上/visual-system/style/style-anchor-board-v1.png`.
- Image 1: `projects/2026-06-23-在路上/visual-system/characters/character-anchor-v2-codex-posture-test.png` as master identity / clothing / prop / boot continuity reference.
- Image 2: `projects/2026-06-23-在路上/visual-system/characters/character-anchor-v2-walk-side-reference-crop-tight.png` as walking side posture, stride, coat movement, backpack placement, and boot profile reference.
- Generation path note: Image 1 and Image 2 are loaded in the visible task context and must be used as references before host image generation.

## Operation Notes

- Task type: new image generation using visible references, not edit target replacement.
- First read order: wet mud and old boot sole, dragging footstep, smeared footprint, then cropped coat / leg / backpack trace.
- The face is not a focus and may be absent or cropped out.
- Composition must avoid commercial product framing even though the boot is close.

## Advisor / Host-Native Record

- Skill mode command: `node C:\Users\wangq\.agents\skills\gpt-image-2\scripts\check-mode.js --json`.
- Skill mode 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 / host-native prompt workflow.
- Garden/API expected: no.
- Template/reference path read: `C:\Users\wangq\.agents\skills\gpt-image-2\references\scenes-and-illustrations\concept-scene.md`.
- Prompt construction also followed host image generation guidance from `C:\Users\wangq\.codex\skills\.system\imagegen\references\prompting.md`.

## Source Files Actually Read

- `AGENTS.md`
- `docs/workflows/slide-task-orchestration.md`
- `docs/workflows/slide-task-templates.md`
- `projects/2026-06-23-在路上/slides/s08/s08-v3-brief.md`
- `projects/2026-06-23-在路上/tasks/2026-06-24-character-consistency/session-06/packets/s08-v3-slide-agent-packet.md`
- `projects/2026-06-23-在路上/project.md`
- `projects/2026-06-23-在路上/execution-plan.md`
- `projects/2026-06-23-在路上/slides/slides.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`
- `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\.codex\skills\.system\imagegen\SKILL.md`
- `C:\Users\wangq\.codex\skills\.system\imagegen\references\prompting.md`

## Final Prompt

Use Image 1 as the master identity / clothing / prop / boot continuity reference. Use Image 2 as the walking side posture / stride / coat movement / backpack placement / boot profile reference. Generate a new 16:9 photorealistic cinematic documentary road-film still.

Low ground close-to-medium-close view on a cold wet muddy road. The first read must be wet mud, rough stones, an old worn boot sole, and a dragging footstep scraping through the mud. The boot is mid-step with weight and fatigue in the motion: heel worn down, tread packed with mud, mud smeared backward, a partial footprint dragged behind it. The traveler body is only a cropped secondary trace: lower dark pants, edge of a long weathered dark coat, maybe a faint backpack strap or coat hem if naturally visible. Preserve the boot shape, heavy old hiking/work-boot feel, dark worn coat, road-worn clothing, and walking posture from Image 1 and Image 2.

The shot should feel like a documentary frame caught from the ground, not staged product photography. Use real muddy road texture, wet dark earth, grit, puddle skin, tiny stones, cold overcast gray light, low saturation cold gray / earth brown / dark khaki / coal black / wet blue-gray palette. Keep the world and road physically present even in the close view. The emotion is bodily cost and persistence: the traveler keeps moving, tired but not collapsed.

Camera and composition: 16:9 horizontal, low ground lens near mud height, close or medium-close framing, boot and dragging footprint dominate the lower and middle frame, any upper body is cropped and secondary. Natural imperfect documentary focus, tactile mud and boot tread detail, no clean studio lighting, no beauty lighting, no heroic stride.

## Negative / Avoid

No shoe advertisement, no product catalog framing, no fashion footwear, no clean sneaker, no polished boot, no logo, no readable label, no watermark, no UI, no text. No face emphasis, no portrait, no heroic centered stance, no motivational poster, no tourism road, no glossy commercial city, no cyberpunk, no neon, no sci-fi. No gore, injury close-up, collapse, or misery exploitation. Do not make the mud decorative or too clean. Do not turn the boot into the only subject detached from the traveler and road.

## Output Intent

- Output file: `projects/2026-06-23-在路上/slides/s08/s08-v3-image.png`.
- Intended use: S-08 production still for the MV image sequence.
- Acceptance target: the image reads first as mud / boot sole / dragging footstep, keeps boot/body continuity from Image 1 and Image 2, and avoids shoe-ad or fashion drift.