/generate-image
Image Generator
Generate branded images via Gemini API
B2B TierContent & Editorial
Overview
End-to-end image generation that builds structured PromptSpec JSON with brand color injection and logo placement, then executes the Gemini CLI to produce actual image files. Handles carousels, blog featured images, social graphics, and newsletter headers.
What It Does
- Builds PromptSpec JSON using the 4-section pattern (Content, Style, Composition, Constraints)
- Injects brand colors by hex+name and prevents typography leakage
- Executes the Gemini CLI to produce .jpg/.png files in standard output directories
- Supports multiple slide types per brand (Title, Data, Quote, Diagram, etc.)
Inputs
- Content source (blog post, key points, headline)
- Brand config
- Image type (carousel, featured, social)
Outputs
- Generated .jpg/.png image files
- PromptSpec JSON
Example
/generate-image
You just finished a blog post and need a LinkedIn carousel. The skill pulls key points, builds 8 slide prompts with your brand colors and logo injection, runs the CLI, and saves the carousel images to the output directory.
Deep Dives
Ready to use /generate-image?
This skill ships with every Knowledge OS installation. Set up your system in 90 minutes.
Built and maintained by Victor Sowers at STEEPWORKS