seo-image-gen creates images for SEO uses through an image generation MCP server (nanobanana, backed by Gemini). Each command maps a use case such as an OG image, blog hero, product photo, infographic or Pinterest pin to an aspect ratio, resolution and domain mode.
After every generation it returns the saved path, the prompt that was sent, the settings and an SEO checklist covering alt text, file naming, WebP conversion and ImageObject or og:image markup.
What You Get
- Generated images saved by the image MCP server, sized for the chosen SEO use case
- The crafted prompt and settings used for each image
- A checklist for alt text, file name, WebP conversion and size targets
- Ready ImageObject and og:image snippets for the new file
Common Use Cases
- Make a 16:9 social preview image for a new landing page
- Produce several blog hero variations and pick one
- Create white-background product shots for a catalogue
- Draft a tall infographic or Pinterest pin
What Makes It Unique
It ties image generation to SEO outcomes. A table of use cases fixes ratio, resolution and style, nine reference files cover prompt writing, safety rephrasing, model limits, cost tracking and presets, and the output always ends with markup and file-handling advice. It checks that the image server is connected before it does anything.
Under The Hood
The skill is instructions plus nine reference files loaded on demand. It first checks that the nanobanana MCP tools are available and tells the user to run the extension installer if they are not. It then calls those tools, which send the prompt to the Gemini API with the user’s key and save the image to disk. The post-generation steps suggest ImageMagick commands that convert and resize files on the machine, and the references describe a cost log and ffmpeg recipes. The frontmatter does not declare shell, network or file access. Fetched content is to be treated as untrusted data.