p-image — Claude Code skill: Pruna image generation
Runs Pruna P-Image models from the terminal: text-to-image, editing existing pictures, LoRA style presets and compositing several images into one.
npx skills add 101-skills/superpowers --skill p-image -a claude-codeWhat p-image does
The skill wires four Pruna models into your agent: pruna/p-image for text-to-image, pruna/p-image-lora with built-in style presets, pruna/p-image-edit for edits and pruna/p-image-edit-lora for edits in a chosen style. Everything goes through the belt CLI from inference.sh, so the agent just runs a command and reads the result.
You need the CLI itself and a belt login before the first run: the skill sits on top of npx skills add belt-sh/cli. Prompts travel as JSON, for example belt app run pruna/p-image --input '{"prompt": "...", "aspect_ratio": "16:9"}'. Aspect ratios are explicit: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, or custom with width and height. The LoRA variant ships presets such as photos-realism, pixel-art, cinematic-movie-style and neon-punk.
Editing takes a list of images in the images field, so you can swap a background or build a collage from several files in one call. Skip belt login and the commands simply fail — check that first. For image work inside a coding session, the skill saves you a trip to the browser.
What it helps with
- generate an image from a text prompt without leaving the terminal
- replace the background on an existing photo
- build a collage from several images in one command
- render an image in a fixed style using a LoRA preset
- set an exact aspect ratio for a layout
Example prompts for the agent
- generate a futuristic cityscape at sunset with p-image
- change the background of this photo to a beach using pruna/p-image-edit
- make a portrait in the photos-realism style with p-image-lora
How to install p-image
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill p-image -a claude-code2Manually: put the files in place
~/.claude/skills/p-image/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/p-image/
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill p-image -a codex2Manually: put the files in place
.agents/skills/p-image/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill p-image -a cursor2Manually: put the files in place
.cursor/skills/p-image/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install 101-skills/superpowers p-image2Manually: put the files in place
.github/skills/p-image/SKILL.md1Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill p-image -a gemini-cli2Platform CLI
gemini skills install https://github.com/101-skills/superpowers --path tools/image/p-imageManually: put the files in place
.devin/skills/p-image/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add 101-skills/superpowers --skill p-image -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 p-image
How do I install p-image in Claude?
Run npx skills add 101-skills/superpowers --skill p-image -a claude-code. The skill lands in ~/.claude/skills/p-image/SKILL.md.
What does p-image do?
It gives your agent commands for generating and editing images with Pruna P-Image models through the belt CLI from inference.sh.
What do I need before using it?
The belt CLI installed and a completed belt login. Without authorization the generation commands will not run.
