DitherStudio

Dither any image in the browser. Clean monochrome studio plus CLI and HTTP API for coding agents.

Studio · DitherSkill · GitLab · GitHub

Live

• Studio: https://ditherstudio.ideatr.dev
• Agent skill + examples: https://ditherskill.ideatr.dev
• Example pack: /examples in the studio (portrait, landscape, UI, logo, gradient)

npx skills add arjunkshah/ditherskill -g -y

Regenerate example gallery

npm run examples

Web studio

npm install
npm run dev

Drop an image, pick an algorithm, export PNG. All client-side.

CLI

npm run cli -- photo.jpg -o out.png -a floyd-steinberg -t 128 --json
npm run cli -- clip.mp4 -o clip-dither.mp4 -a bayer-8 --video-fps 12 --video-crf 24
npm run cli -- algorithms --json

Video input uses local ffmpeg/ffprobe for frame extraction and encoding. Add --no-audio to drop the source audio track.

HTTP API (agents)

npm run serve
# http://127.0.0.1:8787

curl -sS -X POST \
  "http://127.0.0.1:8787/v1/dither?algorithm=atkinson&threshold=140" \
  -F "file=@photo.jpg" -o out.png

• GET /health
• GET /v1/algorithms
• GET /openapi.json
• POST /v1/dither
• POST /v1/dither/base64

Library

import { ditherFile, ALGORITHMS } from 'ditherstudio'

await ditherFile('in.png', 'out.png', {
  algorithm: 'stucki',
  threshold: 128,
})

Algorithms

Threshold · Random · Floyd-Steinberg · Atkinson · Jarvis-Judice-Ninke · Stucki · Burkes · Sierra · Bayer 2/4/8 · Halftone · Blue-noise · Riemersma · Hybrid

Docs for agents

• AGENTS.md
• llms.txt
• Companion skill: arjunkshah/ditherskill

Stack

Vite · React · TypeScript · Tailwind · sharp (CLI/API)

License

MIT