# Generate and Edit Images with AI

Choosing and invoking image models can require repeated CLI research. This skill provides model guidance and ready workflows for generation, editing, upscaling, and stitching.

## Install

```bash
npx skillstore add infsh-skills/ai-image-generation
```

## Metadata

- Status: approved
- Slug: infsh-skills-ai-image-generation
- Skillstore revision: r2
- Version status: missing
- Tree hash: 2cb9ce991233bcec73fba893a5a640197340eaa3a255e3c3c81e31142d880fca
- Author: infsh-skills
- GitHub username: infsh-skills
- License: MIT
- Repository: https://github.com/infsh-skills/skills/tree/main/tools/image/ai-image-generation/
- Ref: 4121de961d1b6f2ffca856260e239505c302452c
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network
- Quality score: 69
- Public page: https://skillstore.pages.dev/skills/infsh-skills-ai-image-generation
- Manifest: https://skillstore.pages.dev/api/skills/infsh-skills-ai-image-generation/manifest

## Capabilities

- Lists image applications available through the belt CLI.
- Generates images from text prompts with supported hosted models.
- Edits existing images from user-provided image URLs.
- Selects models for LoRA styles, text rendering, speed, or image quality.
- Upscales images and stitches multiple images through dedicated applications.

## Use Cases

- Create Campaign Visuals: Generate product scenes, social graphics, and promotional images in selected aspect ratios.
- Explore Visual Concepts: Compare model outputs for characters, environments, posters, and presentation concepts.
- Automate Image Preparation: Edit, upscale, or stitch source images through repeatable CLI workflows.

## Prompt Templates

### Generate a Basic Image

```
Create an image of [subject] in [style]. Use [aspect ratio] and choose a suitable model.
```

### Create a Product Visual

```
Generate a professional product image for [product]. Use [background], [lighting], [camera angle], and [aspect ratio].
```

### Edit an Existing Image

```
Edit the image at [image URL]. Change [requested element] while preserving [details to keep]. Select a model suited to editing.
```

### Compare Model Options

```
Recommend three available models for [goal]. Compare quality, speed, cost, text rendering, and editing support before generating the selected option.
```

## Limitations

- Requires the inference.sh belt CLI and an authenticated account.
- Generation sends prompts and supplied image URLs to remote model services.
- Model availability, pricing, accepted inputs, and output quality can change.
- The skill provides CLI guidance but does not include local image models.

## Best Practices

- Specify the subject, composition, lighting, style, and aspect ratio.
- Review model pricing and input requirements before running generation.
- Use trusted image URLs without embedded credentials or sensitive query parameters.

## Anti Patterns

- Do not send confidential images or prompts without reviewing provider policies.
- Do not assume every model supports editing, LoRA, or identical parameters.
- Do not install related skills unless they are required and independently reviewed.

## Security Audit

- Audited at: 2026-08-06T11:50:00.159\+00:00
- Summary: The reported shell backtick findings are Markdown formatting, not Ruby or shell command substitution. Most URLs are documentation or placeholders, but the remote image can contact an external host when rendered. The broad Bash\(belt \*\) permission remains a medium-risk capability because it permits more CLI actions than this image workflow requires.

## Stats

- Views: 107
- Downloads: 11
- Favorites: 0
- Popularity score: 0