Skip to content
TakoAPI
Back to skills

stability-ai

by 1999azzar

Browser & Automation
625Downloads reported by clawskills.sh
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Generate high-quality images via Stability AI API (SDXL, SD3, Stable Image Core)

Description

What This Skill Does

Generates images using Stability AI's API, supporting models like SDXL 1.0 and Stable Diffusion 3. Each run saves the image locally alongside a JSON metadata file. An auto-cleanup routine keeps only the last 20 generated images on disk.

Provides CLI access to multiple Stability AI models with built-in metadata tracking and seed-based reproducibility, without requiring a separate GUI or third-party image app.

When to Use It

  • Generating concept art for a game or story
  • Creating widescreen wallpapers in 21:9 or 16:9 format
  • Producing reproducible images by reusing a specific seed
  • Making pixel-art or anime-style illustrations from text prompts
  • Generating product mockup backgrounds with a negative prompt to exclude clutter

Example Workflow

INPUT

User asks: Generate a cyberpunk street at night in 16:9 ratio with photographic style

AGENT
  1. 1

    Parses the prompt and options: --ratio 16:9 --style photographic

  2. 2

    Calls the Stability AI API with the prompt and default model (SDXL 1.0)

  3. 3

    Writes the returned image data to a local PNG file

  4. 4

    Saves a JSON metadata file alongside the image recording prompt, model, and settings

  5. 5

    Runs auto-cleanup, removing oldest files if more than 20 images are stored

OUTPUT

Local file path to the generated PNG and a JSON metadata file containing prompt, model, parameters, and generation timestamp

Requirements

  • STABILITY_API_KEY environment variable
  • API_HOST environment variable (optional, for custom endpoints)

Installation

clawhub install 1999azzar/stability-ai

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links