image-poster

Generates single poster and illustration images via the OpenDesign media dispatcher.

Updated May 20, 2026
One-click install
npx skills add https://github.com/Keimoshi/Silverloft --skill image-poster-keimoshi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: image-poster
Source: https://github.com/Keimoshi/Silverloft/tree/main/open-design-port/skills/image-poster
Command: npx skills add https://github.com/Keimoshi/Silverloft --skill image-poster-keimoshi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating a polished poster, key art, or editorial illustration requires assembling a structured image-generation prompt and wiring it to the right model, aspect ratio, and output location. This Skill turns a loose creative brief into a finished PNG/JPEG saved directly into the project folder. ## Core Features & Use Cases - Structured prompt assembly: Builds prompts in a fixed order covering subject, lighting, palette, camera, and negative constraints to avoid common AI-image artifacts. - Project-aware dispatch: Reads the project's imageModel, imageAspect, and imageStyle metadata, then calls the unified media dispatcher instead of raw provider APIs. - Provider-agnostic workflow: Defaults to gpt-image-2 but the same flow drives Flux, Imagen, or Midjourney through the active upstream tooling. - Use Case: Ask for an editorial poster for an indie film festival with a muted ochre palette, and receive a rendered PNG written into the project folder, visible in the FileViewer. ## Quick Start Ask the agent to generate a poster for your project, describing the subject, mood, and palette, and it will produce and save the image file.

Frequently Asked Questions about image-poster

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a poster image with an AI agent?

Describe the subject, composition, lighting, and palette in your brief. The Skill assembles a structured prompt and dispatches it through the media generate command, saving a PNG or JPEG into the project folder.

Which image models does this poster workflow support?

It defaults to gpt-image-2 but is provider-agnostic. The same workflow drives Flux, Imagen, or Midjourney through the active upstream tooling, using the imageModel value stored in project metadata.

Can I generate multiple image variations in one turn?

By default the Skill produces exactly one finished image per turn. Variations are only generated when the user explicitly asks for them, keeping each render deliberate and reviewable.

Does the aspect ratio of the generated image matter?

Yes. The Skill honors the project's imageAspect value exactly, since upstream cost is identical and matching the aspect avoids a wasteful re-render of the same prompt.

Why does the generated poster avoid placeholder text?

The Skill forbids filler typography unless the user requests in-frame text, because image models produce warped lettering. Real copy supplied by the user is used instead of lorem ipsum.