post-image

name	post-image
description	Generate a Gemini image prompt (1200x627 LinkedIn infographic) plus alt text from a finished post draft. Use this AFTER draft-post N when the user wants the visual asset for a specific post — picks one of 6 visual concepts based on the post content.
argument-hint	<post-number>

post-image

Read a completed LinkedIn post draft and generate a detailed image generation prompt optimized for Gemini (e.g., via the nanobanana MCP). Output is a ready-to-use prompt file plus alt text.

When to use

Run after /linkedin-content:draft-post N for the same post number.

User input

$ARGUMENTS

$ARGUMENTS must be a positive integer (the post number).

Argument parsing

• No argument: Output "ERROR: Post number required. Usage: /linkedin-content:post-image N" and STOP.
• Not a valid integer: Output "ERROR: Invalid post number '$ARGUMENTS'." and STOP.

Prerequisites

Verify draft files exist for post N. Check for at least one file matching content/posts/post-N/draft-*.md.

If no drafts exist, output and STOP:

ERROR: Missing prerequisite.
- No drafts found for post [N] at content/posts/post-N/
- Run /linkedin-content:draft-post [N] first to generate them.

Process

Step 1: Load post context

1. Read draft variants for post N in preference order:
   • draft-technical.md (preferred — most concept-rich for visual representation)
   • draft-business.md
   • draft-founder.md
2. Read content/posts/post-N/hooks.md if it exists — hooks often contain the most visually evocative framings.
3. Extract:
   • Core message (1 sentence)
   • Key visual concepts — technical terms, metaphors, or contrasts that can be visualized
   • Target audience — infer from the chosen variant (developer, business, community)

Step 2: Optional brand override

If content/.config.md exists, read brand styling overrides such as:

• Accent color — hex code (default: tech-neutral blue #2563EB)
• Background — default: dark navy / charcoal
• Logo policy — default: no logos (most logos are trademarked)
• Style notes — additional brand guidance

Step 3: Choose visual concept

Pick ONE concept from this priority list — the first that fits the post's core message:

1. Before/After Split — when the post contrasts an old approach vs. a new one
2. Architecture Diagram — when the post explains how components connect
3. Mapping/Translation — when the post draws parallels between two systems
4. Layered Stack — when the post describes layers of a system
5. Flow/Sequence — when the post describes a process from start to end
6. Stats/Metrics Highlight — when the post has specific numbers (e.g., "0 dependencies, 12 lines")

Step 4: Generate the image prompt

Build a Gemini prompt that includes:

1. Format: LinkedIn infographic, 1200x627px landscape format, high resolution
2. Style direction: Clean, modern, professional. Dark background (navy or charcoal) with the accent color from Step 2 (default #2563EB), white text, subtle gradients. No stock-photo look. Technical but accessible.
3. Layout description: Precise placement — left, center, right, top, bottom.
4. Content elements: Exact text labels, icons, or symbols. Keep text minimal — max 6–8 short labels.
5. Visual metaphor: The core concept translated into a visual.
6. Exclusions: No photos of people, no generic tech imagery, no clipart, no code screenshots, no third-party logos (trademarks).

Prompt structure template

Create a LinkedIn infographic (1200x627px, landscape).

**Visual concept**: [1-2 sentences describing the core visual metaphor]

**Layout**:
- [Left/Top section]: [what goes here]
- [Center]: [what goes here]
- [Right/Bottom section]: [what goes here]

**Style**: [color palette, typography direction, mood]

**Text labels** (exact text to include):
- [Label 1]
- [Label 2]
- ...

**Do NOT include**: [exclusions]

Step 5: Write output

Write to content/posts/post-N/image-prompt.md.

Format:

# Image Prompt: Post N — [Topic Title]

**Post**: N
**Source variant**: [technical/business/founder]
**Visual concept**: [chosen concept type from Step 3]
**Generated**: [YYYY-MM-DD]

## Core Message (for visual)

[1 sentence — the single idea the image must communicate]

## Gemini Prompt

[Full prompt text, ready to copy-paste into Gemini / nanobanana]

## Alt Text (for LinkedIn)

[1-2 sentence accessible description of the generated image — for the LinkedIn alt-text field]

## Usage Notes

- **LinkedIn image specs**: 1200x627px recommended for feed posts
- **Fallback**: If generation fails, use the alt text as a text-only post header
- **Regeneration**: Adjust the prompt and re-run if the visual doesn't match the post's tone

Step 6: Output confirmation

After writing the file, output:

Image prompt written to content/posts/post-N/image-prompt.md
Copy the "Gemini Prompt" section into Gemini / nanobanana to generate the infographic.

Error handling

Condition	Response
Missing prerequisites	HARD BLOCK — see Prerequisites
Post number invalid	ERROR with usage instructions
No $ARGUMENTS provided	ERROR with usage instructions
Image generation tool unavailable	Write prompt file, output INFO message with manual instructions
Image generation fails	Keep prompt file, output WARNING with the error, suggest manual generation