GZ-L/photo-to-poster-skills

# Charcoal Fading Travel Study

## Delivery and mode selection

- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — true source photo above, charcoal memory study below.
  2. **Full canvas** — the whole poster becomes a charcoal fading study.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Clean Ink Zen Memory

## Delivery and mode selection

- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — contemplative real photography above, clean ink reconstruction below.
  2. **Full canvas** — the whole poster becomes a clean ink memory study.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Folk Ink Landmark Portrait

## Delivery and mode selection

- Use only for photos with clear architecture or a human-made landmark. If none is recognizable, do not invent one; explain that the image is unsuitable.
- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — real source photo above, folk ink portrait below.
  2. **Full canvas** — the whole poster becomes the folk ink portrait.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected mode guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Gouache Travel Still Life

## Delivery and mode selection

- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer three modes when the user has not already chosen:
  1. **Split still life** — real source photo above, extracted gouache still life below.
  2. **Full-canvas still life** — the entire poster becomes a recomposed object archive.
  3. **Full-scene painting** — preserve the source composition and repaint the complete scene in opaque gouache.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-still-life.md](references/split-still-life.md), [full-canvas-still-life.md](references/full-canvas-still-life.md), or [full-scene-painting.md](references/full-scene-painting.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# 3D Impasto Miniature Landscape

## Delivery

- Use the source photo only as a reference for subject, structure, pose, spatial relationships, and color. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately; never create a collage.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Preserve subject identity, count, and defining features. You may refine angle, proportions, and local structure for the miniature format, but the result must remain immediately traceable to the source.

## Fixed visual specification

Reconstruct the source image's most recognizable subject, silhouette, pose, and narrative relationship as a refined, dimensional, tactile 3D impasto miniature landscape.

- Use bright white textured paper, generous negative space, and a centered or slightly off-center diagonal axis.

# 3D Isometric Healing Blocks

## Delivery

- Use the source photo only as a reference for subject, structure, pose, spatial relationships, and color. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately; never create a collage.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Preserve subject identity, count, mass proportions, axes, curves, interlocks, layers, occlusions, and negative spaces. Simplify secondary detail without changing defining features.

## Fixed visual specification

Use a deconstruct → select → distill → reconstruct process to translate the subject into a healing-style 3D/isometric block sculpture. Reorganize it with rounded, simple, tactile modular volumes while keeping it immediately recognizable.

- Adapt the composition to the subject's center of gravity, pose direction, and spatial relationships.

# Kraft Stencil Landmark

## Applicability, delivery, and mode selection

- Use only for a clearly isolated landmark or object with a strong single silhouette, such as a lighthouse, tower, gateway, monument, windmill, pavilion, bridge tower, or lone boat. If no single subject is recognizable, do not impose a stock landmark.
- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — true source photo above, kraft stencil print below.
  2. **Full canvas** — the whole poster becomes a kraft stencil print.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Memory Diary Duet

## Delivery

- Generate one standalone 4:5 vertical poster per source image. Never combine multiple photos.
- Divide the canvas with a diagonal or near-vertical line into complementary areas: real photography on the left and a hand-drawn interpretation of the same moment on the right. Keep the boundary clear, elegant, and aligned with the subject; no gradients, feathering, or media blending.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when none is available.

## Left photograph

Preserve subject identity, count, structure, pose, object relationships, environment, materials, natural light, and viewpoint. Apply only subtle lifestyle-editorial grading with warm natural light, soft contrast, faint nostalgia, and fine grain. Do not alter subjects, add or remove objects, replace the setting, invent regional details, or illustrate the photo.

## Right memory drawing

# Nostalgic Travel Editorial

## Delivery and mode selection

- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — nostalgic real photography above and a minimal drawing below.
  2. **Full canvas** — the complete source scene becomes a nostalgic editorial illustration.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Observational Pen and Wash

## Delivery

- Use the source photo only as a reference for subject, structure, pose, spatial relationships, color, and mood. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately; never create a collage.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Preserve subject identity, count, proportions, direction, defining pose, and scene meaning. Simplify surroundings and secondary details without distorting or replacing defining features.

## Fixed visual specification

Extract only the source image's most recognizable subject, contour, pose, and narrative relationship. Keep the visual clues worth recording, using the minimum information required for immediate recognition while preserving evidence of observation, omission, and stopping.

- Treat the core contour or action as a visual motif. Organize the full page through repetition, variation, pause, accent, and sudden negative space.

# Paper Cutout Travel Figures

## Applicability, delivery, and mode selection

- Use only for photos with recognizable people. If no person is clear enough to preserve, do not add figures; explain that the image is unsuitable.
- Produce one standalone 3:4 vertical poster per source image. Never combine photos.
- Offer two modes when the user has not already chosen:
  1. **Split comparison** — full source photo above, reconstructed real environment with paper figures below.
  2. **Full canvas** — the whole poster keeps the real environment while all recognizable narrative people become paper figures.
- Use one mode for all uploaded photos unless the user maps modes per image.
- Read only the selected guide: [split-comparison.md](references/split-comparison.md) or [full-canvas.md](references/full-canvas.md).
- Return a ready-to-use prompt only when explicitly requested; otherwise use the available image generation or editing capability.

## Shared visual rules

# Handcrafted 3D Travel Diorama

## Delivery

- Use the source photo only as a reference for subject, structure, pose, spatial relationships, scene, and color. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately; never create a collage.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Preserve subject identity, count, silhouette, pose, proportions, direction, defining structures, spatial relationships, and narrative context so the result remains immediately traceable to the source. Do not distort, replace, or invent defining features.

## Fixed visual specification

Reconstruct the subject and only the few elements that best reinforce its identity and setting as a refined 3D handcrafted miniature scene. Use paper, cardstock, soft clay, or thin wood. Simplify details without mechanically copying the photograph.

- Build the composition around one dominant subject, one long narrow lightweight floating landscape base, and generous negative space.

# Pixel Dissolve Landscape

## Delivery

- Use the source photo only as a reference for subject, structure, space, direction, and color. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately; never create a collage.
- If the user asks only for a prompt, return a ready-to-use generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Create a manually selected, color-limited, recomposed pixel landscape—not a mosaic filter or globally downsampled photo.

## Fixed visual specification

On warm off-white or pale ivory paper, choose one highly recognizable visual anchor from the source, such as a bridge, facade, dome, vehicle, boat, gate, or roof group. Retain only the minimum elements required to communicate subject identity, place, and spatial relationships.

Preserve subject identity, count, pose, architectural proportions, road direction, occlusions, viewpoint, and perspective. Remove repetitive windows, tile texture, tiny signage, tourist clutter, dense cables, foliage noise, and photographic noise when they are nonessential. Never alter defining structures or invent landmarks, people, vehicles, or scene elements.

# Watercolor Travel Card

## Delivery

- Use the source photo only as a reference for subject, structure, space, direction, and color. Stylize the entire final canvas; do not show the original photo or create split-screen or before-and-after layouts.
- Generate one standalone 3:4 vertical poster per source image. Process multiple images separately and never combine them in a single generation.
- If the user asks only for a prompt, return a ready-to-use image-generation prompt. Otherwise, use the available image generation or editing capability directly. Ask for an upload only when no reference image is available.
- Before generation, lock the subject's identity, count, pose, proportions, direction, architectural or terrain structure, spatial relationships, and viewpoint. Do not invent a location, date, coordinates, or attribution.

## Fixed visual specification

The entire canvas is a perforated watercolor travel memory card on warm ivory cotton cold-pressed paper, divided vertically:

- The left 68%–75% reconstructs the scene with transparent watercolor and a few fine pen lines. Preserve the most recognizable contours, pose, direction, colors, and spatial relationships. Remove clutter, repeated objects, micro-texture, and irrelevant detail without changing defining features.