OpenClaw .NET Image Generator

Utility script that uses ElBruno.Text2Image.Foundry v0.7.0 to generate all OpenClaw .NET brand images from the prompt catalog at docs/design/image-prompts.md using FLUX.2 Pro on Microsoft Foundry.

Supports image-to-image (img2img) — select prompts use the official .NET logo as a reference image for brand-consistent generation.

Prerequisites#

.NET 8.0 SDK
A deployed FLUX.2 Pro model on Azure AI Foundry

Setup#

Store your FLUX.2 Pro credentials securely with .NET user-secrets (one-time):

cd scripts\ImageGenerator
dotnet user-secrets set Endpoint "https://your-resource.services.ai.azure.com"
dotnet user-secrets set ApiKey "your-api-key"
 
# Optional overrides:
dotnet user-secrets set ModelId "FLUX.2-pro"     # deployment/model name
dotnet user-secrets set ModelName "FLUX.2 Pro"   # display name

Alternatively, environment variables also work: FLUX2_ENDPOINT, FLUX2_APIKEY, FLUX2_MODELID, FLUX2_MODELNAME.

Usage#

# List all available prompts
.\generate.ps1
 
# Generate ALL images (saves to both repos)
.\generate.ps1 all
 
# Generate specific image(s) by ID
.\generate.ps1 1A 2A 3B
 
# Generate all images in a category
.\generate.ps1 category:logo
.\generate.ps1 category:slides
.\generate.ps1 category:blog
.\generate.ps1 category:social
.\generate.ps1 category:github
.\generate.ps1 category:webapp
 
# Preview what would be generated
.\generate.ps1 --dry-run all

Output Locations#

Images are saved to both repositories:

Repo	Path
`openclawnet-plan`	`docs/design/assets/{category}/{filename}.png`
`openclawnet`	`docs/design/assets/{category}/{filename}.png`

Image Categories#

Category	IDs	Description
`logo`	1A, 1B, 1C, 1D	Logo/mascot variants
`slides`	2A–2E	Presentation slide backgrounds
`blog`	3A–3F	Blog/article header images
`social`	4A–4D	Social media cards
`github`	5A–5C	GitHub social previews & README banner
`webapp`	6A–6C	Blazor web app assets