1.0.1
Fixed
- Running scripts/prompt.mjs through a symlinked install now works. Before, it exited without writing anything.
image-prompt
Write image prompts in the project's brand, sized for each platform, ready for any image generator. Reads the project's own brand file for palette, look, tone and a don't list, and never stores brand values itself. Writes the prompt, a JSON with size, ratio and a suggested model, and on request short paste steps for a web tool. It never generates an image and never spends credits. Needs Node 18 or newer.
Say this you do not type commands, you ask
- “write an image prompt”
- “prompt for an Instagram post image”
- “make a prompt for a product photo”
- “give me a prompt I can paste into ChatGPT”
- “brand prompt for this product”
- “what size should this image be”
- “image prompt”
Once it is installed, that is the whole interface. Your agent picks the skill up on its own and runs whatever it needs to. The commands further down are there for anyone who would rather drive it themselves.
Install pick your agent
Changelog what changed, newest first
1.0.0
Added
- Write an image prompt in the project's brand, from a brand file the project owns. No brand file is fine.
- Four directions: editorial, clean studio, detail and an overlay with room for a headline.
- Exact sizes for product pages, Instagram, Facebook, LinkedIn, X, Pinterest and YouTube thumbnails.
- A JSON beside each prompt with the size, ratio, don't list and a suggested Higgsfield model, with the crop needed when a model cannot make the exact ratio.
- Short paste steps for any web image tool.
- A prompt log per project that stops the same prompt being written twice.
An agent that has this installed reads CHANGELOG.md in the skill folder.
The same history is at changelog.json,
and every tool's releases are at /updates.json.
Just ask
Once the skill is installed, say what you want in your own words:
"write a prompt for an Instagram post of our new tote bag" "make it a close-up of the stitching, for a story" "give me a prompt I can paste into ChatGPT, with our product photo as reference" "write a product page image prompt, but ignore the brand this time"
The agent writes the prompt, shows it to you, and says where it saved it. Nothing is generated and nothing is spent. To make the image, hand the prompt to an image generator: the higgsfield-studio skill when it is installed, which prices every job first, or any web tool by paste.
Why this exists
A good image prompt carries a lot: the brand's colours and look, the right size for the platform, what to avoid, and a line that stops the model from handing a reference photo back unchanged. Writing that by hand every time is slow, and it drifts. This skill builds it the same way every time, from the brand file the project already keeps, and keeps a log so the same prompt is not written twice.
Requirements
Node 18 or newer. If you do not have it, the skill offers to install it:
sh scripts/setup-deps.sh --check # is everything present?
sh scripts/setup-deps.sh # show the install command, ask, then run it
.\scripts\setup-deps.ps1 # Windows
It works offline and needs no account.
The brand file
Keep a brand.md in the project (or pass --brand FILE). Sections are found by their headings: palette, photography, studio, motion, tone and don't. The full contract and an example are in references/brand-file.md. Without one, prompts carry no brand style.
If you prefer the command line
node scripts/prompt.mjs sizes
node scripts/prompt.mjs brand
node scripts/prompt.mjs --subject "a tan leather tote bag" --platform instagram_post --variant portrait --people --paste ChatGPT
Each run writes prompt.txt, prompt.json and, with --paste, PASTE.md to a dated folder under .prompts/ (or --out DIR). .prompts/log.jsonl is the log the duplicate check reads. Keep it.
Built at Itqan Lab, a design and technology studio.
إتقان — itqan, the Arabic word for mastery: doing a thing precisely, and completely.