Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated the usage example to save intermediate and output files into the `/tmp` directory instead of the current directory. - Improved script hygiene by avoiding local file clutter for generated images and results.
What This Skill Does
Generates images through an Azure Foundry (Cognitive Services / OpenAI) deployment and returns the result as raw PNG/JPEG bytes or a URL. Requires a configured Foundry endpoint, API key, and deployment name.
Wraps Azure Foundry's image generation REST API into a single, environment-variable-driven skill, avoiding manual curl construction in each workflow.
When to use it
- Generate product mockup images from text prompts
- Create visual assets for automated content pipelines
- Batch-produce illustrations for internal tools
- Test Azure AI image model deployments via CLI
- Integrate image generation into OpenClaw agent workflows
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: generate a 1024x1024 PNG of a red fox
- 1Validates FOUNDRY_ENDPOINT format to reject malformed URIs
- 2Constructs the deployment URL from FOUNDRY_ENDPOINT, FOUNDRY_DEPLOYMENT, and FOUNDRY_API_VERSION
- 3Sends a POST request with the prompt payload using curl and the FOUNDRY_API_KEY header
- 4Decodes the base64 image from the JSON response and writes it to /tmp/generated_image.png
PNG file saved at /tmp/generated_image.png
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.