ai-product-photography — Claude Code skill: product photos via belt
ai-product-photography turns a product brief into commercial images through the inference.sh CLI: studio shots, lifestyle scenes and marketplace cards.
no license 243 installs ★ 31 updated 2026-08-31
npx skills add 101-skills/superpowers --skill ai-product-photography -a claude-codeWhat ai-product-photography does
The skill routes image requests to inference.sh through the belt CLI. The agent picks a model, assembles a prompt from product, background, lighting, style and technical detail, then runs the app. Model IDs available: falai/flux-dev for detail, falai/flux-schnell when you want speed, google/imagen-3, xai/grok-imagine-image, bytedance/seedream-3-0.
Recipes cover the usual e-commerce setups: white studio background, lifestyle context, hero angle, flat lay, action shot. Categories get their own examples, from electronics and apparel to food, furniture and jewellery. Amazon main-image and Shopify banner templates are included. A bash loop produces several angles of one item; falai/topaz-image-upscaler takes scale 2, falai/birefnet cuts the background out.
Nothing runs until the inference.sh CLI is in place: npx skills add belt-sh/cli installs the belt skill, then belt login authorises the session. Every recipe calls belt app run, so an unauthenticated run fails on the first command. The frontmatter allowlist is Bash(belt *), which keeps the agent inside that one binary.
What it helps with
- shoot a white-background listing image for Amazon
- render several angles of one product in a batch
- build a lifestyle hero for a Shopify banner
- upscale a generated shot and strip its background
- pick a model for speed versus fine detail
Example prompts for the agent
- Generate a studio shot of wireless earbuds on a white background for an Amazon listing
- Make a lifestyle photo of a coffee mug on a wooden desk in morning light
- Run four angles of a luxury watch with
falai/flux-dev, then remove the background
How to install ai-product-photography
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill ai-product-photography -a claude-code2Manually: put the files in place
~/.claude/skills/ai-product-photography/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/ai-product-photography/
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill ai-product-photography -a codex2Manually: put the files in place
.agents/skills/ai-product-photography/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill ai-product-photography -a cursor2Manually: put the files in place
.cursor/skills/ai-product-photography/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install 101-skills/superpowers guides/photo/ai-product-photography2Manually: put the files in place
.github/skills/ai-product-photography/SKILL.md1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill ai-product-photography -a gemini-cli2Platform CLI
gemini skills install https://github.com/101-skills/superpowers --path guides/photo/ai-product-photographyManually: put the files in place
.devin/skills/ai-product-photography/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill ai-product-photography -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about ai-product-photography
How do I install ai-product-photography in Claude Code?
Run npx skills add 101-skills/superpowers --skill ai-product-photography -a claude-code. The skill lands at ~/.claude/skills/ai-product-photography/SKILL.md.
What does ai-product-photography do?
It writes prompts and calls inference.sh models to produce product, lifestyle and advertising images, with batch angles and post-processing included.
What do I need before using it?
The inference.sh belt CLI: npx skills add belt-sh/cli installs it and belt login authorises the session. Without that login the generation commands fail.