Images

On-brand images from text, using your colors, logos, and fonts.

The Images superskill lets your AI agent generate images that match your brand identity. It automatically injects your brand's logos, color palette, and fonts as context for the AI model, so every generated image feels on-brand without manual image editing work. Choose from multiple aspect ratios and provide reference images for style-guided generation. All images are stored in a searchable gallery for easy reuse.

What your agent can do

Capabilities available out of the box.

  • Generate images from text prompts with automatic brand context injection
  • Use your brand logos, color palette, and fonts as visual references
  • Choose from multiple aspect ratios for social posts, stories, and banners
  • Provide reference images for style-guided generation
  • Browse and manage generated images in a searchable gallery

Integrations

Platforms and services your agent connects to.

Gemini 3 ProGemini 3 Pro

Commands

CLI commands available with this superskill.

Generate Image

Generate an image from a text prompt. Brand styling is applied automatically unless --no-branding is used.

clawpow images create
Edit Image

Edit an existing image using AI. Provide a source image and edit instructions.

clawpow images edit
Upscale Image

Upscale an image to 4x resolution using AI super-resolution.

clawpow images upscale
Get Image

Get image generation details, including status and result URL.

clawpow images get <imageId>
List Images

List generated images with filters for status.

clawpow images list
Skill Reference
# Images

Turn ideas into on-brand visuals instantly -- generate, edit, and upscale images from the CLI.

## When to Use
- User needs a social media graphic, blog hero image, or marketing banner
- User wants to edit an existing image (remove background, change text, adjust colors)
- User wants to upscale a low-resolution image to higher quality
- User needs on-brand visuals for a campaign or launch

## How It Connects
- **Branding** -- Brand colors, logos, and fonts are automatically applied to generated images. Set up branding first for consistent visuals.
- **Social Media** -- Generated images can be attached to social posts using the returned asset ID with the `--media` flag.
- **Blog** -- Use generated images as hero images or inline visuals in blog articles.
- **Video** -- For moving content, use the video skill instead. Images handles static visuals only.

## Quick Start

Generate an image:
```bash
clawpow images create --prompt "A hero banner for our product launch" --json
```

## Commands

### Generate Image
```bash
clawpow images create --prompt "Description of the image" --json
clawpow images create --prompt "Description" --aspect-ratio 16:9 --json
clawpow images create --prompt "Description" --no-branding --json
clawpow images create --prompt "Description" --reference-images assetId1 assetId2 --json
```
Generate an image. Brand styling is applied automatically unless --no-branding is used.

**Options:**
- `--prompt <text>` (required) -- Text description of the desired image
- `--aspect-ratio <ratio>` -- Aspect ratio: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 (default: 1:1)
- `--reference-images <assetId...>` -- Asset IDs of reference images for style guidance
- `--no-branding` -- Skip brand styling

**Returns:** `{"imageId": "...", "status": "generating"}`

### Edit Image
```bash
clawpow images edit --prompt "Remove the background" --image <assetId> --json
clawpow images edit --prompt "Change the text to say Hello World" --image <assetId> --json
clawpow images edit --prompt "Make the colors warmer" --image <assetId> --no-branding --json
clawpow images edit --prompt "Crop to landscape" --image <assetId> --aspect-ratio 16:9 --json
```
Edit an existing image using AI. Provide a source image asset ID and edit instructions. Brand styling is applied automatically unless --no-branding is used.

**Options:**
- `--prompt <text>` (required) -- Edit instructions describing the desired changes
- `--image <assetId>` (required) -- Asset ID of the source image to edit
- `--aspect-ratio <ratio>` -- Aspect ratio: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 (default: 1:1)
- `--no-branding` -- Skip brand styling

**Returns:** `{"imageId": "...", "status": "generating"}`

### Upscale Image
```bash
clawpow images upscale --image <assetId> --json
```
Upscale an image to 4x resolution using AI super-resolution. No prompt or branding options needed -- just provide the source image.

**Options:**
- `--image <assetId>` (required) -- Asset ID of the image to upscale

**Returns:** `{"imageId": "...", "status": "generating"}`

### Get Image
```bash
clawpow images get <imageId> --json
```
Get image generation details and the result asset URL when available.

**Options:**
- `<imageId>` (required) -- Image generation ID

**Returns:** `{_id, prompt, aspectRatio, status, resultAssetId, resultUrl, _creationTime}`

### List Images
```bash
clawpow images list --json
clawpow images list --status ready --json
```
List generated images with optional filters.

**Options:**
- `--status <status>` -- Filter by status: generating, ready, failed

**Returns:** Array of `{_id, prompt, aspectRatio, status, resultAssetId, resultUrl, _creationTime}`

## Output Format

All commands support `--json` for structured output. Without `--json`, output is human-readable.

## Important Notes

- Always use `--json` for machine-parseable output
- Image generation, editing, and upscaling are paid operations (consume credits)
- Image listing and detail lookup are free operations
- Generated images are stored as assets and can be used in social posts
- Supported aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9

Related superskills

Works great together with these.