deck-image-prompts

Generates image prompts for six AI engines to fill presentation storyboard slides.

1|Updated Apr 19, 2026
One-click install
npx skills add https://github.com/fercosnt/fernando-claude-marketplace --skill deck-image-prompts-fercosnt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deck-image-prompts
Source: https://github.com/fercosnt/fernando-claude-marketplace/tree/main/deck-builder/skills/deck-image-prompts
Command: npx skills add https://github.com/fercosnt/fernando-claude-marketplace --skill deck-image-prompts-fercosnt

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve? Writing effective image-generation prompts for presentation slides requires knowing each engine's syntax, negative prompt handling, and brand consistency tactics. This Skill produces ready-to-paste prompts for slide storyboards while enforcing a whitelist of which slide types should get images at all. ## Core Features & Use Cases - Multi-engine prompt generation: Produces 3 prompt variations per slide for Midjourney v7, Google Imagen 4, Nano Banana Pro, Higgsfield Soul, DALL-E 3/GPT-Image-1.5, and Ideogram 3.0, each with aspect ratio, composition, and negative/constraints fields. - Whitelist enforcement (D5 rule): Automatically fills prompts for cover, problem, conceptual, comparison, demo, and social-proof slides, and returns an explicit skip for financial, data, CTA, and disclaimer slides. - Brand and compliance guardrails: Injects design-system color tokens, bans stock-photo clichés, applies anti-uncanny rules for people, and enforces medical compliance (ANVISA laser, TCLE disclaimers) for clinical content. - Use Case: After generating a fundraising pitch storyboard, invoke this Skill to receive three engine-specific image prompts for the cover slide with brand palette hex codes embedded, while the financial slide is correctly marked as skipped. ## Quick Start Ask the assistant to generate image prompts for the cover slide of your pitch deck using Midjourney, Imagen 4, and Nano Banana Pro.

Frequently Asked Questions about deck-image-prompts

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

FAQPage Schema
How do I generate image prompts for presentation slides?

Provide the slide type, key message, visual brief, and brand, and the Skill returns three prompt variations for different engines. Each variation includes aspect ratio, composition direction, the prompt text, and negative or constraints fields ready to paste into the image tool.

Which image engine should I use for slide images?

Use Ideogram 3.0 when text must appear inside the image, Nano Banana Pro for 4K or 21:9 output and brand consistency, Midjourney v7 for hero and lifestyle imagery, and Higgsfield Soul for recurring trained characters. The Skill defaults to Higgsfield, Imagen 4, and Nano Banana Pro.

Does Midjourney v7 support negative prompts?

Yes, Midjourney v7 supports negative prompts via the --no parameter with weight -0.5, where each comma-separated term is read independently. Imagen 4 and GPT-Image-1.5 lack native negative fields, so restrictions must be rewritten as positive constraints inside the prompt.

Why does the Skill skip financial and data slides?

Financial and data slides are skipped because charts and tables require numeric precision that image generators cannot deliver. The Skill returns an explicit skip marker so the slide uses a real chart or table instead, unless the user passes an include flag or the scientific override applies.

How do I keep image style consistent across all slides in a deck?

Use the same style anchor across the series: a fixed --sref code and seed in Midjourney, a trained Soul ID in Higgsfield, a style_code in Ideogram, or a verbatim style description in Imagen 4. The Skill applies these anchors automatically when multiple slides are processed.

Can AI image prompts be used for medical or clinical slides?

Yes, but the Skill enforces compliance guardrails: it describes positive serene scenes first, bans blood, needles, and distress imagery, and prohibits visual promises of cures. Before-after comparisons require a TCLE notice and a results-may-vary disclaimer.