In Claude, open Customize, then Connectors. Click + Add, then Add custom connector. Paste https://mcp.higgsfield.ai/mcp and sign in. Then Claude can make pictures for you. It uses paid credits. Free pictures don't work through the plug.
What it does: Reads your past Higgsfield images and their prompts, finds the looks you use, and writes one style guide and a base prompt so new pictures match.
Use it when: Your AI pictures look different every time and nobody knows which look is the brand.
You get: Your looks grouped with counts, the one to keep, a base prompt, a never list, and a saved skill.
Open in
See the prompt
You are my brand designer. You can read my Higgsfield history.
1. Read my last 30 images and videos: the prompt, the model, the size and the date. Tell me how many you read.
2. Group them into looks (for example: photo style, flat drawing, lo-fi art, app icons). For each look: how many, the dates, the colors named in the prompts, and one short example prompt line.
3. Tell me which look I use most lately, and where looks clash (two color sets for the same brand).
4. Write a style guide for the main look: light, colors (only hex codes found in my prompts), camera or drawing style, mood, and what's always in it (for example: "no text, no logos").
5. Write one base prompt under 50 words I can start every new picture from.
6. Write a never list with every avoid line in my prompts (no text, no logos, and so on), and say which look each came from.
7. If you can't tell whether two looks are for the same brand, write [check: same brand?].
Rules:
- Only what's in my history. Don't guess colors or rules that aren't in my prompts.
- Don't make or change any images.
- Use only what is in my data and what I tell you. Never make up a name, number, date, price, link or promise.
- 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 'Picture Style Guide' so you do it this way each time I make a new picture.
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 from: Extreme Ownership, Jocko Willink and Leif Babin, Chapter 6: Simple
What it does: Writes the Higgsfield prompt and picks a camera move that makes your real product photo move without changing the product.
Use it when: You have good product photos but your ads and posts need video.
You get: A video prompt, the camera move to pick, what must not change, and 3 versions to try.
Open in
See the prompt
You are my video prompt writer. I'll use Higgsfield's image to video.
My product: [what it is]
The photo: [describe it: angle, background, light]
Where it will run: [for example: Instagram Reels ad, 9:16]
The feel: [for example: clean, premium, calm]
1. Pick one camera move that fits (for example: a slow push in, a slow orbit) and say why in one line.
2. Write the video prompt, under 60 words: what moves (light, steam, a hand), what stays still, the feel.
3. Write the "must not change" list: the product's shape, logo, label words, colors.
4. Write 3 versions to try: one with only the camera moving, one with a small motion in the scene, one with light changing.
5. A check list for each result: is the label still readable, did the shape change, no extra hands or fingers, no strange text.
Rules:
- Keep the product true. A video that changes the product can't be used in an ad.
- Use only what is in my data and what I tell you. Never make up a name, number, date, price, link or promise.
- 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 'Photo To Video' so you do it this way each time I make a video from a product photo.
Built from: $100M Leads, Alex Hormozi, Section III: Get Leads, Run Paid Ads Part I: Making An Ad
What it does: Reads your product page and writes 3 ad ideas with a hook, a scene and a Higgsfield prompt for each, so you can test which one works.
Use it when: You want to test video ads but don't know what to make first.
You get: 3 ad ideas with a 3 second hook, the scene, the prompt, the on-screen words, and which to test first.
Open in
See the prompt
You are my ad planner. I'll make the videos in Higgsfield.
My product page text: [paste]
Who buys it: [who]
What they really want: [the feeling or result, in their words]
Write 3 ad ideas, each 5 to 10 seconds, each a different angle:
- A: the problem, then the product as the fix
- B: the product in use, close up
- C: the result they want, shown, not said
For each:
1. The hook for the first 3 seconds, under 8 words, on screen.
2. The scene in 2 lines.
3. A Higgsfield prompt under 60 words.
4. On-screen words for the end, under 8 words. Every ad ends with the same call to action.
Then say which to test first and why, in one line.
Rules:
- Only claims from my product page. No "best" or numbers I didn't give.
- No people who look like real famous people.
- If you have to pick something I didn't give you (a name, a time, a setting, a day), mark it (suggested).
- Use only what is in my data and what I tell you. Never make up a name, number, date, price, link or promise.
- 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 'Ad Angles' so you do it this way each time I plan video ads.
Built from: Cashvertising, Drew Eric Whitman, Chapter 1: What People Really Want
Simple AI
MarketingRun the business
