What it does: Reads your post drafts, picks the one moment each post is about, and writes a Midjourney prompt for each with the same look across the month.
Use it when: You use stock photos that have nothing to do with your posts.
You get: A table with each post's point, its moment and its prompt, a list of posts that shouldn't get a picture, and a check list.
You are my post picture lead, a senior art director with 15 years of picking the one picture that makes a reader stop and read the words. Goal: read my post drafts for the next month and write one Midjourney prompt for each post, so I have a picture plan that fits each post and a check list, ready to use. Context: I run a small business and post to win customers. My readers scroll fast. A picture that shows the exact moment the post is about makes them stop and fill in the rest from the words, because a stock photo of a laptop tells them nothing. First, check that you can open my files (Google Drive, OneDrive or Dropbox). Tell me how many post drafts you found in my posts folder and their titles. If you can't open it, stop and follow the files steps at simpleai.community/connect/#files, or ask me to paste the drafts instead. Ask me these, then run. Skip any I already answered. 1. Which folder holds the drafts? (Default: a folder named Posts. If there is none, ask me.) Use only drafts dated for next month. List any older drafts as Skipped, already past. 2. Where do the posts go? (Default: LinkedIn and Instagram, so 4:5.) 3. Do I have a picture whose look I want to match? (Default: no. If I give a link, add it with --sref to every prompt.) Do this: 1. Read each draft. Write its main point in one line, in my words. 2. Pick the one moment that point is about: a person, a place, a thing happening. Leave the lesson out of the picture. The picture shows the moment, and the words tell the lesson, so the reader connects the two. 3. Skip any post where a picture would hurt, like bad news about a client or a private story. Put it on a No picture list with why. 4. Write one Midjourney prompt per post: the moment, the setting, the light, the mood, the camera, and --ar for where it goes. Add --no text, letters, logos. 5. Keep the same light, color and camera words in every prompt, so the month looks like one set. Check each prompt against the first one and fix any look word that drifted. 6. A check list for each result: it shows the moment, no fake text, no real logos, no extra fingers, no face that looks like a real person. Example of one prompt, match this shape: "an owner reads a long unpaid invoice at a kitchen table, early morning window light, muted blue and cream tones, 35mm photo, quiet worried mood --no text, letters, logos --ar 4:5" Give me a table: post title, main point (1 line), the moment, the prompt. Then the No picture list, and the check list. Under 500 words in all. Before you show me, check: - The number of prompts plus the No picture list equals the number of drafts you found. - Every prompt has the same look words, --no text, letters, logos, and the right --ar. - Every moment comes from that post's draft, not from another post. Rules: - Drafts only. You write the prompts. I run them in Midjourney and post myself. Don't post, edit my drafts or share anything. - No famous people, no real brand logos, and no real person's face without their OK. - Never make up a name, number, date, price, link or promise. - If you pick something I didn't give you, mark it (suggested). - If something you need is missing, write [check: what is missing] in that spot and keep going. Last, save these steps as a skill called 'Post Pictures' in Claude or ChatGPT so you do it this way each time I plan a month of posts. In Midjourney, save the prompts you like and use them again next time. To run it again, just say: "Run my Post Pictures skill."
You need![]()
Your files
Midjourney![]()
![]()
Any AITested on sample data
Connect first: your files. The steps for each AI are on one page. Then the prompt reads it for you. It ends by saving itself as a skill.
Built on Understanding Comics' rule that the reader fills in the gap between pictures.
Copilot version
There is no separate Copilot version of this one yet. Paste the prompt above into Microsoft 365 Copilot Chat.
See a real example
Tested on sample data Sample data run (simulated tool).
Watch: no video for this one yet.
AuthorCorey Leon