# Flux Image Generation

> State-of-the-art AI image generation using Black Forest Labs FLUX models Use when this capability is needed.

- Skill: `tomevault-io/flux-image-generation` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add tomevault-io/flux-image-generation`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/flux-image-generation/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: tomevault-io (https://skillmd.com/u/tomevault-io)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tomevault-io/flux-image-generation

---


# FLUX Image Generation

State-of-the-art AI image generation using Black Forest Labs' FLUX models.

## Overview

This skill provides access to FLUX, one of the most advanced AI image generation models available. FLUX excels at photorealistic imagery, complex compositions, and accurate text rendering in images.

## Available Tools

### Text-to-Image

- {{flux_pro}}: High-quality image generation with FLUX Pro 1.1
- {{flux_flex}}: Flexible generation with adjustable guidance scale

### Image Editing

- {{flux_kontext}}: Context-aware image editing and transformation
- {{flux_fill}}: Inpaint/fill masked regions of images
- {{flux_expand}}: Outpaint/extend images beyond boundaries

## Key Features

### Superior Quality

FLUX Pro 1.1 represents the cutting edge of image generation, with:

- Exceptional photorealism
- Accurate text rendering in images
- Complex multi-subject compositions
- Consistent style and coherence

### Context Understanding

FLUX Kontext understands image context to make intelligent edits:

- Object modification while preserving surroundings
- Style transformation with content preservation
- Seamless blending of edited regions

## Usage Examples

### Example 1: Generate High-Quality Image

Use {{flux_pro}} for photorealistic generation:

```json
{
  "prompt": "Professional product photo of a smartwatch on marble surface, studio lighting",
  "width": 1024,
  "height": 1024,
  "prompt_upsampling": true,
  "output_format": "png"
}
```

### Example 2: Edit with Context

Use {{flux_kontext}} for context-aware transformations:

```json
{
  "prompt": "Change the season from summer to winter, add snow to the ground and trees",
  "guidance": 3.5
}
```

**Blob inputs:** `image` (summer photo to edit)

### Example 3: Expand Canvas

Use {{flux_expand}} to outpaint image boundaries:

```json
{
  "prompt": "Continue the background scenery naturally",
  "top": 256,
  "bottom": 0,
  "left": 128,
  "right": 128
}
```

**Blob inputs:** `image` (portrait to expand)

### Example 4: Fill Region

Use {{flux_fill}} with a mask to fill specific areas:

```json
{
  "prompt": "Beautiful sunset sky with orange and purple clouds",
  "guidance": 3.5,
  "prompt_upsampling": true
}
```

**Blob inputs:** `image` (source photo), `mask` (sky region marked white)

## Parameters

### Common Parameters

| Parameter | Description | Default |
|-----------|-------------|---------|
| prompt | Text description of desired output | Required |
| width/height | Output dimensions (256-1440) | 1024 |
| seed | Random seed for reproducibility | Random |
| output_format | png or jpeg | png |

### Generation Parameters ({{flux_pro}}, {{flux_flex}})

| Parameter | Description | Default |
|-----------|-------------|---------|
| guidance | Guidance scale (1.5-5) | 3.5 |
| prompt_upsampling | Auto-enhance prompts with details | false |
| safety_tolerance | Content filter level (0-6) | 2 |

### Expand Parameters ({{flux_expand}})

| Parameter | Description | Default |
|-----------|-------------|---------|
| top | Pixels to expand at top | 0 |
| bottom | Pixels to expand at bottom | 0 |
| left | Pixels to expand at left | 0 |
| right | Pixels to expand at right | 0 |

## Best Practices

1. **Prompt Upsampling**: Enable for more detailed results with {{flux_pro}}
2. **Guidance Scale**: Lower (2-3) for creative freedom, higher (4-5) for prompt adherence
3. **Safety Tolerance**: Adjust based on content needs (0=strict, 6=permissive)
4. **Seed Usage**: Save seeds to reproduce or iterate on good results

## Important: Language Requirements

**⚠️ All prompts MUST be in English!**

FLUX APIs only support English prompts. If the user provides a prompt in another language (e.g., Chinese, Japanese, Spanish), you MUST:

1. Translate the prompt to English
2. Enhance with descriptive keywords while preserving the original intent

Example: "可爱的男孩遛狗" → "A cute young boy walking a friendly dog in a sunny park, warm atmosphere, heartwarming scene, high quality"

## Limitations

- Async processing with polling (may take 10-60 seconds)
- Maximum dimensions: 1440x1440 pixels
- Some content filtered by safety systems
- API rate limits apply
- **Prompts must be in English only**

---
> Converted and distributed by [TomeVault](https://tomevault.io/claim/shazhou-ww) — claim your Tome and manage your conversions.
<!-- tomevault:4.0:skill_md:2026-04-14 -->