Skip to content

smixs/visual-skills / image

Image prompting skill for Nano Banana (NBP/NB2) and GPT Image 2.5 (Flare/Sunburst). Writes ready-to-use prompts with model/quality/size recommendations. Use when: "нарисуй", "сгенерируй картинку", "image prompt", "промпт для картинки", blog covers, slides, posters, product shots, UI mockups, storyboards, character sheets, edit/colorize, style transfer, vision analysis, image-to-prompt, nb, NBP, NB2, gpt-image-2.5, multi-panel grids, ecommerce product photography, fashion editorial, food/beverage ads, cinematic portraits. Do NOT use for: video (use video skill), 3D models, audio, non-image tasks.

references/vision-decomposer.md

View on GitHub
# Vision Decomposer — Image-to-Prompt Analysis

Use this reference when the user asks to analyze an image and convert it to a generation prompt: **style transfer**, **mood reference**, **"сделай промпт по этой картинке"**, **"перенеси стиль/образ"**, **"проанализируй кадр"**, **"image to prompt"**, **"reverse-engineer this look"**.

You become a **professional Vision Agent**. Your task is deep cinematic, psychological and optical-colorimetric analysis of the image, translating it into a perfect, highly detailed text prompt for generation.

The process has two strict steps. Do not skip Step 1. Do not freestyle Step 2.

---

## STEP 1 — DEEP DECOMPOSITION

Scan the image as **raw data** and extract facts based on:

- **Cinematic composition** — Bruce Block's visual structure
- **Color theory** — Itten
- **Perceptual psychology** — Arnheim
- **Directing references** — Spielberg / Scorsese / Tarantino
- **Cinematography & colorimetry** — Valentin Zheleznyakov

Evaluate the following four parameter blocks.

### 1. Subject — psychology and mise-en-scène

- **Who / What:** precise clothing (era, style, materials), age, skin texture, makeup / toning, micro-expressions.
- **Shot size (after Mascelli):** Extreme Long Shot (ELS), Long Shot, Medium Shot, Two-shot / Three-shot, Close-Up (CU), Choker CU, Extreme CU, Insert shot.
- **Directing & force placement (Blocking after Kenworthy):** isolating the character at the edge of the frame (anxiety), physical barriers, face turned away from camera, shooting from behind (effect of unknown / powerlessness). Height difference (Level Change) for dominance.
- **Kinetics & perceptual forces (Arnheim):** visual weight of the object, center of gravity. Internal tension of form (compression / stretching / twisting). Pose, plasticity, motion vectors. Imitation of swift action (motion blur) or monumental stillness.
- **Color and emotion (Itten):** psychological impact of the local color of the subject.
- **Gaze:** direction, eyeline anchor, interaction with the frame edges, look into the lens.

### 2. Environment — geometry and structure of space

- **Setting & Art Direction:** place, era, background, architecture, specifics of materials and surface textures (gloss, matte, rust).
- **Three-plane depth (Spielberg):** clear separation into Foreground (with leading details), Midground (action plane), Background.
- **Figure & ground (Arnheim):** degree of subject isolation, overlapping of forms, mass relationships. Use of reflections (mirrors, windows) to expand context.
- **Tonal & aerial perspective (Zheleznyakov):** drop in contrast, desaturation, washing out, shift toward cool (blue-cyan) tones in the background.
- **Frame geometry:** structural framework (axes of symmetry / asymmetry), claustrophobic compression, surface divisions. Illusion of depth via Foreground framing.

### 3. Lighting — chiaroscuro, color and contrast

- **Optical & visual contrast (Zheleznyakov):** ratio of lit and shadow areas (OVK). Depth and density of shadows. Exact local color vs. color modified by lighting (valeurs).
- **Lighting scheme:** Key, Fill, Backlight / Rim, Modeling. Rembrandt lighting, Chiaroscuro. Height, hardness (Hard / Soft) and type of source.
- **Broken light & reflexes (Zheleznyakov):** use of shadow masks (gobos), light through blinds / foliage, color reflexes from neighboring objects onto skin / clothing.
- **Tonality & color:** light key (High-key / Low-key). Threshold Silhouette. Itten's 7 contrasts. White balance (Daylight / Tungsten) and temperature contrast (warm light / cool shadows).

### 4. Tech & Cinematography — optics, filters and film texture

- **Camera (angles & viewpoint):** Angle (High angle, Low angle, Eye-level, Dutch tilt). Objective camera, Subjective, POV, Over-the-shoulder (OTS). Movement imitation (Push-in, Tracking shot).
- **Optics & DOF:** Focal length (wide-angle for distortion / coverage, telephoto for spatial compression). Depth of field. Rack focus, bokeh.
- **Optical filters & attachments (Zheleznyakov):** diffusion filters (Pro-Mist, Black Pro-Mist, Fog, Double Fog, Low Contrast) for highlight bloom (halation), skin softening, lowering of micro-contrast. Polarizers (cutting reflections).
- **Effects & texture:** Frame format (70mm, 35mm, IMAX). Exposure (motion blur). Stylization (Bleach Bypass). Film Grain, chromatic aberration.

---

## STEP 2 — PROMPT SYNTHESIS

Assemble the final prompt from Step 1 data.

Rules:
- Write **only keywords separated by commas, in English**.
- Describe **strictly what you see**. Do not invent new objects.
- **No filler.** Never write "The image shows...", "A picture of...", "I can see...".
- **Strict word order** (this formula is non-negotiable):

```
[Shot type, optics and angle], [Subject, mise-en-scène / blocking, visual weight, clothing and action], [Multi-plane environment (Foreground / Midground / Background), overlapping and geometry], [Lighting scheme, optical contrast, gobos and reflexes], [Color palette, temperature contrast and aerial perspective], [Color Grading, diffusion filters, film stock, textural artifacts]
```

---

## OUTPUT PROTOCOL

When the user gives you an image (or asks to reverse-engineer one), produce the answer in **two blocks**:

### Block 1 — Brief Analysis Log

```
Subject & Blocking: ...
Environment & Depth: ...
Lighting & Contrast: ...
Tech & Optics: ...
```

Each line: 1-2 sentences max. Just the extracted parameters.

### Block 2 — Final Prompt

A code block containing only the prompt text in English, assembled by the formula in Step 2:

````
```
<comma-separated keywords following the strict 6-segment formula>
```
````

After the prompt, append the standard image-skill output header (Model / Quality / Size) so the user can drop it straight into the generator.

---

## When to load this file

Load `vision-decomposer.md` whenever:
- The user attaches an image and asks for a prompt to **recreate / transfer / match** its style
- The user asks to **decompose, deconstruct, reverse-engineer** a visual reference
- The user says **"перенеси стиль", "сделай так же", "повтори образ", "проанализируй кадр", "разбери картинку на промпт"**
- Mood-board work: extracting cinematic DNA from a film still, ad frame, painting, photo

Do NOT load this for: pure generation requests where no reference image is given. For those, use the standard model files.

---

*Author: Serge Shima ([t.me/aimastersme](https://t.me/aimastersme) · [sergeshima.com](https://sergeshima.com) · [aimasters.me](https://aimasters.me)) · License: CC BY 4.0 — attribution required · Source: [smixs/visual-skills](https://github.com/smixs/visual-skills)*
canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill.Stars: 198kUpdated Claude CodeCodexVerifiedInstall