stable-diffusion

Stable Diffusion image generation models. Use for image AI.

7 stars

byG1Joshi

View on GitHub Installation ↓

Best use case

stable-diffusion is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Stable Diffusion image generation models. Use for image AI.

Teams using stable-diffusion should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/stable-diffusion/SKILL.md --create-dirs "https://raw.githubusercontent.com/G1Joshi/Agent-Skills/main/skills/ai-ml/stable-diffusion/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/stable-diffusion/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How stable-diffusion Compares

Feature / Agent	stable-diffusion	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

Stable Diffusion image generation models. Use for image AI.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Stable Diffusion

Stable Diffusion (by Stability AI) is the open standard for image generation. SD 3.5 (2025) improves prompt adherence and typography.

## When to Use

- **Control**: You need exact composition control (ControlNet).
- **Local Generation**: Run on your own GPU. No censorship/cost.
- **Fine-Tuning**: training LoRAs on your own face/product.

## Core Concepts

### Diffusers Library

The Hugging Face library to run SD pipelines in Python.

### LoRA (Low-Rank Adaptation)

Small adapter files (100MB) that add a style or character to the base model.

### ComfyUI

The node-based GUI for building complex SD workflows (see separate skill).

## Best Practices (2025)

**Do**:

- **Use SD 3.5 Large**: For best text rendering.
- **Use Flux**: The community has largely moved to **Flux.1** (Black Forest Labs) alongside SD.
- **Use ControlNet**: To force the image to follow a specific pose or edge map.

**Don't**:

- **Don't use SD 1.5**: Unless you need specific legacy LoRAs. SDXL/SD3/Flux are superior.

## References

- [Stability AI Models](https://stability.ai/models)