Create image

Skill sujithq/skills/skills/create-image

Collection of skills

Install

npx -y skills add sujithq/skills --skill create-image

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Generate images using Azure OpenAI GPT Image models. Use this skill when the user asks to create images, generate visuals, or produce artwork using AI.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.1 KB, 724 tokens by cl100k_base, as published. Nobody here has run it

Image Generation with Azure OpenAI

This skill generates images using Azure OpenAI GPT Image models through direct API calls.

When to use

User asks to generate, create, or produce an image
User wants to visualize a concept or idea
User requests artwork, illustrations, or graphics
User asks to create visual representations of descriptions

Prerequisites

User must have:

Azure OpenAI resource with deployed GPT Image model (gpt-image-2 or gpt-image-1.5)
Azure authentication configured:
- Recommended: Azure CLI (az login) with "Cognitive Services OpenAI User" role
- Alternative: API key (if enabled on the endpoint)

Important: Many Azure OpenAI endpoints disable API key authentication. Always use Azure RBAC when API keys are disabled.

How to use

Using Python

Install required packages:

pip install openai azure-identity

Generate image with Azure RBAC (recommended):

from openai import AzureOpenAI
from azure.identity import DefaultAzureCredential, get_bearer_token_provider

token_provider = get_bearer_token_provider(
    DefaultAzureCredential(),
    "https://cognitiveservices.azure.com/.default"
)

client = AzureOpenAI(
    api_version="2024-02-01",
    azure_endpoint="https://YOUR-RESOURCE.openai.azure.com/",
    azure_ad_token_provider=token_provider
)

result = client.images.generate(
    model="gpt-image-2",
    prompt="A serene mountain landscape at sunset",
    size="1536x1024",
    quality="medium"
)

print(result.data[0].url)

Parameters

prompt: Description of the image (1-4000 chars, required)
model: gpt-image-2 or gpt-image-1.5 (use deployment name)
size: Use model-supported dimensions such as 1024x1024, 1024x1536, or 1536x1024 (default: 1024x1024)
quality: low, medium, high, or auto for GPT Image models (use medium by default)
n: Number of images (model-specific limits apply)
style: Optional stylistic control if supported by the deployed model

Prompt crafting tips

Be specific and descriptive
Include style, mood, and atmosphere details
Specify key visual elements and composition
Use portrait (1024x1536) for vertical, landscape (1536x1024) for horizontal images
Use high quality only when the user specifically requests extra detail; use low for faster or lower-cost drafts

Output

The API returns a temporary URL valid for 24 hours
Download and save the image if persistence is needed
The response includes a revised_prompt showing what the model actually used

Error handling

401/403: Check Azure authentication and RBAC role assignments
API Key Disabled: Switch to Azure RBAC with DefaultAzureCredential
429: Rate limit reached, implement retry with backoff
Content filtered: Revise prompt to comply with content policies

References

Azure OpenAI Image Generation Docs