mira-image-prompt

Generates structured JSON prompts for luxury product photography with AI image models.

Updated Sep 2, 2026
One-click install
npx skills add https://github.com/mayconspirlandelli/slides-podcast-ia-home-office --skill mira-image-prompt-mayconspirlandelli
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mira-image-prompt
Source: https://github.com/mayconspirlandelli/slides-podcast-ia-home-office/tree/main/.agents/skills/mira-image-prompt
Command: npx skills add https://github.com/mayconspirlandelli/slides-podcast-ia-home-office --skill mira-image-prompt-mayconspirlandelli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Writing effective prompts for AI image generators requires knowing the right structure, vocabulary, and technical parameters. This Skill removes the guesswork by guiding you through a structured interview and producing a complete, model-ready JSON prompt for cinematic product photography. ## Core Features & Use Cases - Guided 3-round interview: Collects product details, composition, lighting, colors, and technical specs in organized rounds instead of overwhelming you with questions. - Standardized JSON schema: Outputs a complete master_prompt object covering product, composition, lighting, color palette, technical specs, and output specs (resolution, aspect ratio, model). - Optimized for Nano Banana 2: Prompts are tailored for Gemini 3.1 Flash Image via Google Antigravity, with native resolution (512px to 4K) and aspect ratio parameters. - Use Case: A marketer needs a luxury Instagram photo of a strawberry shake. The Skill interviews them about the bottle, splash action, marble surface, and pink palette, then outputs a ready-to-paste JSON prompt in 1:1 aspect ratio. ## Quick Start Ask the assistant to create an image prompt for your product, for example: "Create a product photo prompt for a chocolate lava cake with a dramatic cinematic style."

Frequently Asked Questions about mira-image-prompt

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

FAQPage Schema
How do I create a JSON prompt for AI product photography?▼

Answer the guided interview covering your product, scene, composition, lighting, and colors. After you confirm the summary, the Skill outputs a complete master_prompt JSON you can paste directly into the image model.

What image models does this prompt format work with?▼

The JSON is optimized for Nano Banana 2 (Gemini 3.1 Flash Image) via Google Antigravity, whose output_specs fields are interpreted natively. The descriptive content can also inform prompts for Midjourney, DALL-E, or Flux.

What resolutions and aspect ratios are supported?▼

Resolutions include 512px, 1K, 2K, and 4K. Aspect ratios include 16:9, 1:1, 9:16, 4:3, 3:4, 4:1, 1:4, 8:1, and 1:8, covering social feeds, stories, banners, and print.

Can I use this for product types other than food and drinks?▼

Yes. The interview and schema work for any visual product including cosmetics, jewelry, fashion, and electronics. The luxury cinematic style and frozen-motion composition rules apply across all product categories.

Why does the generated JSON always include a synthid_watermark field?▼

The synthid_watermark field is always set to true because it is a mandatory default for Google's image models. It ensures generated images carry Google's SynthID provenance watermark.