/infographic

name: infographic
description: Generate professional infographics using Nano Banana MCP (Gemini AI image generation). Follows a guided flow — analyze content, suggest visualizable concepts, propose visualization approaches, then generate on-brand images. USE THIS SKILL WHEN user says "create infographic", "make a visual", "design an infographic", "generate infographic", wants to visualize a concept, framework, or data for social media, mentions "LinkedIn post", "visual content", "infographic series", asks to explain a concept visually or create a diagram-style image, mentions "visual metaphor", "iceberg diagram", "pyramid", "funnel visualization", wants to generate images with Gemini for informational content, or says "brand visuals", "social media graphics", "thought leadership visual".
  - User mentions "nano banana", "gemini image generation", "professional infographics"
disable-model-invocation: true

Infographic Generation Skill

Create professional infographics and editorial illustrations using Gemini AI via the Nano Banana MCP server. This skill follows a strict interactive process — never skip ahead without user approval at each step.

**Core philosophy:** The quality of the output is determined by the quality of the thinking BEFORE any prompt is written. Spend 70% of the effort understanding what to visualize and 30% on how to visualize it.

---

When This Skill Loads (DO THIS IMMEDIATELY)

1. Read `references/design-principles.md` (brand philosophy, content analysis, style discovery — internalize deeply) 2. Read `references/gemini-image-api.md` (MCP tool reference + prompt vector system) 3. Read `references/api-setup.md` (API key setup guide) 4. Execute Phase 0 (silent config check) 5. **If API key is NOT configured in Phase 0:**

• Skip directly to Phase 4 (API Key Setup)
• Do NOT proceed to Phase 1
• Do NOT ask about content

6. **If API key IS configured:**

• Begin Phase 1

---

Phase 0: Silent Config Check

Run silently — do NOT ask questions yet.

0.1: Ensure Directory Structure

mkdir -p .infographic/images
mkdir -p .infographic/prompts

0.2: Check for API Key

Check if the GEMINI_API_KEY exists. Check BOTH the environment variable AND the `.env` file:

# Check environment variable first, then .env file
if [ -n "$GEMINI_API_KEY" ]; then
  echo "API key found in environment"
elif [ -f .env ] && grep -q "^GEMINI_API_KEY=.\+" .env; then
  echo "API key found in .env file"
else
  echo "API key not configured"
fi

**Why check both?** The MCP server uses `dotenv` to load `.env` automatically at startup. So even if the bash shell doesn't have the env var set, the MCP server reads it from `.env` directly. If the key exists in `.env`, the MCP server has it.

**CRITICAL DECISION POINT:**

• If found (in env var OR `.env` file): Proceed silently to Phase 1
• If NOT found in either: **IMMEDIATELY jump to Phase 4 (API Key Setup)**
• Do NOT proceed to Phase 1
• Do NOT ask about content
• Do NOT try to generate
• Go directly to Phase 4 to get the API key from the user

**Note:** Do NOT use `get_configuration_status` MCP tool - it gives false positives.

0.3: Check for Brand Config

if [ -f .infographic/brand.md ]; then
  cat .infographic/brand.md
fi

If brand config exists, acknowledge briefly:

I found your saved settings. Ready to create your infographic!

If nothing exists, proceed to Phase 1.

---

Phase 1: Deep Content Interrogation

**Goal:** Understand the content at the DEEPEST level — not what happened, but what it CLAIMS is true about the world. This is the most important phase. If you get this wrong, everything downstream is wasted effort.

1.1: The Four-Layer Analysis

Read the user's content. Then extract meaning at ALL four layers:

LAYER 1 — NARRATIVE: What happened?
(Events, timeline, biography, sequence of actions)
→ Almost never the right thing to visualize

LAYER 2 — THEMES: What is it about?
(Growth, struggle, identity, change, ambition)
→ Better, but still abstract and leads to generic metaphors

LAYER 3 — CLAIMS: What does it SAY is true about the world?
(Universal truths, philosophical statements, hard-won insights)
→ This is where editorial illustrations live
→ Example: "When the suffering is something you chose, you can work with it.
   When it's something that happened to you, it just eats you."

LAYER 4 — INFORMATION: What concrete data, comparisons, or frameworks exist?
(Numbers, scorecards, before/after, checklists, structures)
→ This is where information graphics live
→ Example: "Developer: 2 strengths, 4 gaps. Domain Expert: 6 strengths, 0 gaps."

Present your analysis explicitly labeled by layer. Be honest about which layers are strong and which are weak for this particular content.

1.2: The Design Brief Question

Use `AskUserQuestion`:

question: "What should a viewer GET from this visual in 2 seconds?"
header: "Intent"
options:
  - label: "Feel something"
    description: "An emotional hit — recognition, discomfort, inspiration. They stop scrolling because it resonates."
  - label: "Learn something"
    description: "A concrete fact, comparison, or framework. They save it because it's useful."
  - label: "Both"
    description: "An emotional hook with concrete information backing it up."

This answer determines EVERYTHING downstream:

• "Feel something" → Editorial Illustration path
• "Learn something" → Information Graphic path
• "Both" → Hybrid path

1.3: Platform

Use `AskUserQuestion`:

question: "Where will this be posted?"
header: "Platform"
options:
  - label: "LinkedIn (Recommended)"
    description: "4:5 portrait, optimized for feed engagement"
  - label: "Instagram"
    description: "1:1 square or 4:5 portrait"
  - label: "Twitter/X"
    description: "16:9 landscape for timeline"
  - label: "Presentation"
    description: "16:9 landscape for slides"

---

Phase 1.5: Output Mode Decision

**Goal:** Lock in the output