image Skill for Claude: Marketing Images, Prompts and Optimization

Makes Claude choose and brief the right way to make a marketing image, from AI models to Canva to web optimization.

Installs: 91.1K; on skills.sh
Stars: 57.9K; on GitHub
Files: 2; besides SKILL.md
Updated: Oct 2026; last repo commit

image makes Claude act as a marketing image producer. It asks about your goal, approach and technical context, then helps you choose between AI generation, AI editing, design tools, screenshot overlays and stock photos. It compares seven image models, covers Canva and Figma, and gives workflows for blog heroes, social graphics, product mockups and profile banners.

It also covers web image optimization and OG and social preview images, and asks whether you have API keys for tools such as Gemini, Replicate or Flux, or Ideogram. The skill is mostly guidance, with a few example shell commands for converting and checking images. The folder holds an AI image prompting reference and no scripts.

Good for

Marketers who need blog, social and listing visuals
Founders without a designer
Developers optimizing images and OG previews

Install image

Claude Code

Copy this prompt.

Install the image skill in Claude Code. Run this command:

npx skills add coreyhaines31/marketingskills --skill image -a claude-code

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

The skills CLI route takes image alone. The plugin installs the whole marketing-skills set, which the repo's marketplace.json describes as 53 skills.

From the project README. Source.

What's in the skill

When it loads

When the user wants to create, generate, edit, or optimize images for marketing — blog heroes, social graphics, product mockups, profile banners, listing visuals, or brand assets. Also use when the user mentions 'AI image generation,' 'generate an image,' 'create a graphic,' 'product mockup,' 'hero image,' 'social media graphic,' 'banner image,' 'cover photo,' 'profile banner,' 'listing screenshot,' 'Flux,' 'Flux Kontext,' 'Midjourney,' 'DALL-E,' 'GPT Image,' 'ChatGPT Images,' 'Ideogram,' 'Gemini image,' 'Nano Banana,' 'Recraft,' 'Stable Diffusion,' 'Canva,' 'Figma,' 'image optimization,' 'compress images,' 'WebP,' or 'OG image.' Use this for general-purpose marketing image creation and optimization. For paid ad image creative and platform-specific ad specs, see ad-creative. For video production, see video.

Sections in SKILL.md

Before Starting
1. Image Goal
2. Production Approach
3. Technical Context
Choosing Your Approach
AI Image Generation
Model Comparison
When to Use Which
Prompting Basics
Design Tools
Canva
Figma
When to Use Design Tools vs. AI Generation
Marketing Image Workflows
Blog & Article Hero Images
Social Media Graphics

Key facts

Author: Corey Haines (coreyhaines31)
Repository: coreyhaines31/marketingskills
Installs on skills.sh: 91,109 (crawled 2026-11-02)
GitHub stars: 57,855 (whole repo)
License: MIT
Last repo commit: 2026-10-31
Files in the skill folder: 3 files: SKILL.md, evals/evals.json and references/ai-image-prompting.md. No scripts.
skills.sh audits: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
Works in (per README): Claude Code, OpenAI Codex, Cursor, Windsurf and other Agent Skills agents

What it includes

Asks about the image goal, production approach and technical context
Before starting
AI generation, AI editing, design tools, screenshot plus overlay, or stock
Choosing your approach
7 models: Gemini, Flux, Ideogram, ChatGPT Images, Midjourney, Recraft, Stable Diffusion
AI model comparison
Notes on Canva and Figma, and when to use them instead of AI
Design tools
Blog heroes, social graphics, product mockups and profile banners
Marketing image workflows
Example cwebp, mogrify and jpegoptim commands and a curl check
Runs code · Image optimization commands
Required meta tags and dynamic OG image options
OG and social previews
Prompting guidance for AI image models
Read · references/ai-image-prompting.md

From the official docs

Permissions

Low risk

Its example commands convert and compress image files in the folder where they run. Check which files a pattern like *.png will match before you approve one.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Example commands run on file patterns
The optimization section shows cwebp, mogrify and jpegoptim on globs such as *.png and *.jpg, and a curl command that fetches a page you name and greps its image tags. Run them in a copy of your images folder.
The plugin route installs the whole set
/plugin install marketing-skills adds every skill in the repo (53 per the marketplace file), not only image. Use the skills CLI command to take image alone.
It asks about image-tool API keys
SKILL.md asks whether you have keys for Gemini, Replicate or Flux, and Ideogram. Keep keys in your own environment and out of chat.

Technical details

Talks to	The page you name, if you use the curl check
Scopes	Image tool API keys you choose to use (Gemini, Replicate/Flux, Ideogram)

Questions

What does the image skill do?

It makes Claude help you produce marketing images. Claude picks an approach, compares AI models and design tools, writes prompts, plans workflows for common asset types and gives optimization steps.

How do I install it?

Use the install box at the top of this page. The README says the skills CLI may install only to .agents/skills/ when run from inside an agent session, which Claude Code does not read, so pass -a claude-code there. It also documents a Claude Code plugin, a Codex plugin and Claude web ZIPs.

Does it generate images itself?

SKILL.md does not bundle an image generator. It guides which tools to use and asks whether you have API keys for Gemini, Replicate or Flux, or Ideogram.

Do I need the other marketing skills?

No. SKILL.md has Claude read .agents/product-marketing.md first if it exists (the .claude/ and older file names also work), then ask only what is missing. The README's product-marketing skill is the one that creates that file. It names ad-creative, video, social, cro, seo-audit, aso and directory-submissions as related skills.

Is it safe?

It bundles no scripts. It includes example commands for cwebp, mogrify, jpegoptim and curl that run on files in the current folder, such as *.png, so check the folder before approving them.

How is it different from the video and social skills?

Its Related Skills section sends video production to video and what to post to social. image covers the visual asset itself.

Information

updated Nov 2, 2026

Repository: coreyhaines31/marketingskills
Folder: skills/image

coreyhaines31/marketingskills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 3

marketingskills README · github.com
marketplace.json · github.com
image SKILL.md · github.com