Hire an AI product photo ad variation planner

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A plan for the set that says what repeats, what changes and how many pieces
• A ready-to-paste image prompt for each piece, with exact words and a fix for each fault
• A check of the set against your proof and ad rules, with a verdict
• Upscale settings and a change check for the images you pick

How it does the job well

• Start from the user problem and the evidence for it, then the solution.
• Write decisions down with the reason and the options you rejected.
• Keep scope small enough to ship and learn. Name what is out of scope.
• Define how success will be measured before building.

Mistakes it avoids

• Treating opinions or one loud request as proven demand.
• Specs that skip edge cases, error states or dependencies.
• Roadmaps with dates but no owners or confidence levels.

What a good result looks like

An example, not your task: You have one clean photo of your product and no budget for a shoot.

• Step 1: Set plan. What repeats, what changes, how many pieces, and the colour for each piece.
• Step 2: Prompts. One image prompt per piece, with every word in quotes, tool settings and a fix for each common fault.
• Step 3: Check. Findings on the prompts against your proof and ad rules, versions that are too alike, and a verdict.
• Pause: make and pick images. You paste the prompts into your image tool and approve the images you want.
• Step 4: Upscale. Upscale settings for the size you need, a change check on your picks, and the full-size image.
• Hand-back summary. Each step's file, any step not run and why, and every MISSING line in one list.

How the work is checked

• Only the colour changes between pieces, and the product shape is the same in all of them.
• Every word in a prompt appears once, in quotes, in reading order.
• The product facts in the prompts match your photo, including the label text.
• The check names any two ads that are the same idea instead of passing them.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Andy Warhol would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Plan the set. Picks what buyers already know on sight, uses your one photo as the only source and lays out the set so only the colour changes.
• Write the exact-text prompts. Turns each piece in the plan into a prompt. Every word is in quotes, in order, with the product facts repeated so it stays true.
• Check before you make them. Reads the prompts against your proof and the ad rules, checks that each version changes one thing and gives a verdict.
• Pause to make and pick images. You paste the prompts into your image tool, set size and quality there, and approve the images you want.
• Get the picks to full size. Sets a faithful upscale for your use, protects small words and logos, and checks that nothing changed.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id repeat-one-product-photo-in-new-colours-as-ads only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/repeat-one-product-photo-in-new-colours-as-ads/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• One clear photo of your product (more angles help)
• The words each ad should carry, or your offer and price
• Your brand file or your colours and voice (optional)
• Product facts such as size and label text (optional)
• The size you need the final images (optional)
• The image tool and the upscaler you will use (step 4 asks which, and an image tool may bill each picture)

Tools and access

• Any good chat AI: Planning the set and steps is text work any strong chat AI does well.
• ChatGPT, Gemini or an image editor: These make or recolour the pictures once the plan is ready.
• GPT Image: Reads numbered photos and quoted words closely, and edits one part while keeping the rest.
• Nano Banana: Takes several product and people photos at once, and Google suggests it for finished images with text.
• Ideogram: Lets you set each block of text and where it goes, and on paid plans you can fix one wrong word without a new image.
• FLUX: Matches hex colour codes tied to each object, which helps when brand colours must be exact.
• Claude: Reads full-size images closely, so it can spell-check the words in the picture and spot the safe zones.
• Meta Ads Manager: Its safe zone guardrail marks the area the app's buttons and labels won't cover, so you see what might get hidden.
• Topaz Gigapixel: Lets you choose a faithful model, including one made for text and logos, and turn face recovery down or off. It is paid software.
• Magnific: Offers a Precision upscaler with plain sharpen and grain sliders, apart from its Creative one. It is a paid service.
• GIMP: Free editor whose Difference layer mode shows any change between the original and the result.

Functions and deliverables

• Product photo ad variation planner: Share one exact product photo and brand guidance. Get matching ad concepts, exact-text prompts, preflight findings and size checks for approved images.
  • A plan for the set that says what repeats, what changes and how many pieces
  • A ready-to-paste image prompt for each piece, with exact words and a fix for each fault
  • A check of the set against your proof and ad rules, with a verdict
  • Upscale settings and a change check for the images you pick
• Repeated image campaign planner: Turn one product photo into look-alike images that change only colour. Get the colour list, steps a helper can follow and a price.
  • The one thing your buyers know on sight, and why
  • The set: how many pieces, with one colour for each
  • Numbered steps a helper can follow to make the next piece
  • A keep-or-fix rule for slips and happy accidents
  • A price for the volume with the sum shown, and a thumbnail check
• Exact-text image ad prompt writer: Share an approved image plan and references. Get a prompt with locked wording, tool settings and a checklist for reviewing generated results.
  • The settings to choose in your image tool, kept apart from the prompt
  • One prompt, ready to paste, with a job for every photo
  • What to check letter by letter when the image comes back
  • A fix for each common fault, one edit at a time
  • A six-point match check, for the prompt and for each image that comes back
• Pre-launch ad checker: Share your ad, audience and approved facts. Get located issues, fixes, human checks and a custom score before launch.
  • A version-bound list of observed problems and the smallest fixes
  • The buyer argument checked against the supplied facts and proof
  • A clear list of media, destination, tracking or spec checks still needed
  • A scoped verdict and observed/eligible coverage
  • An internal editing score, with any post-fix score marked provisional
• Image enlargement quality checker: Check an image upscale before it goes to print. Give it your approved image and target size. You get safe settings, a change check and a score.
  • The real target size in pixels, worked out from where the image will go
  • The kind of upscaler to use and the settings that keep it faithful
  • A plan for keeping your words and logo exact
  • A size check, a difference check and a review of words, faces, hands, product and texture
  • A verdict, either accepted or the original returned with the reason
  • A fidelity score out of 100 for each try, from the share of checked areas that match the original
• Product context file writer: Share product facts, audience and positioning. Get a reviewed context file that other marketing skills can reference.
  • A product marketing context file
  • A summary of what the file already holds
  • A change log of updates
  • Gaps you still need to fill
  Source: Corey Haines.
• Andy Warhol: Plan a bold image or a series of designs using Andy Warhol’s recorded approach. Repeat one familiar subject and vary the colour.
  • Your problem worked on the way Andy Warhol would: their questions, their method, their push-back
  • The answer or the work, in Andy Warhol's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You have one clean photo of your product and no budget for a shoot.
• You want ads that look like one family and are easy to add to later.
• Your last AI images had garbled words or a product that did not look like yours.

Limits to check before starting

• You have no clean photo of your product: take one first, or use the Product image ad prompt writer to describe the product in words.
• You want a single ad, not a matching set: use the Product image ad prompt writer.
• Your product is mostly small label text and you need every word perfect: check the images by eye and fix the words in a simple editor.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Download the handoff pack. It is a zip with HANDOFF.md and every skill file this workflow uses.
• Give the folder to Claude, in Claude Code or as files in a Claude Project, and say "Follow HANDOFF.md".
• Attach your product photo and say what the ads must say.
• When it pauses, paste each prompt into your image tool, such as ChatGPT or Gemini. Set the size and quality there, then pick the images you approve.
• Attach the picked images and ask for the last step.
• Check the full-size images by eye before you publish them.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/repeat-one-product-photo-in-new-colours-as-ads for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.