bedardandy

Coloring Book Skill — AI skill for Claude Code

AI community

Claude Code skill: personalized printable kids' coloring books — parametric characters, story mode for ages 3-6, shape cookbook, vector PDF output (AGPL).

How to install Coloring Book Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open bedardandy/coloring-book-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Coloring Book Skill does

Claude Code skill: personalized printable kids' coloring books — parametric characters, story mode for ages 3-6, shape cookbook, vector PDF output (AGPL).

Alternatives in AI

  • Claude Cookbook — Official notebooks and recipes for common patterns (RAG, tool use, Skills, MCP) 35.3k ★
  • Pixel Agents — Compressed Reference — VS Code extension with embedded React webview: pixel art office where AI agents (Claude Code terminals) are an 6.5k ★
  • Hermes Agent Orange Book — Hermes Agent 从入门到精通 · 橙皮书系列 · Nous Research 开源 AI Agent 框架实战指南 4.9k ★

README

coloring-book — a Claude Code skill for personalized kids' coloring books

Give it your kids' names, hair, glasses, freckles, pets, and favorite things — get back a printable, letter-size vector PDF coloring book starring them. Themed pages ("their favorite things") or **story mode** for ages 3-6 (a gentle 10-beat quest with captions built for pre-readers: simple present tense, name repetition, a chant-along refrain).

*The bundled example, "Where Is Button?" — a fictional family's teddy-bear hunt, built end-to-end by the skill.*

Why it works (design choices)

  • Parametric characters, not AI likeness. Each child is a small trait vector (hair silhouette, glasses, freckles, height, a personal motif). Same recognizable kid on every page, zero identity drift, no photos needed or wanted.
  • Everything is bold vector line art (SVG → PDF), tuned to age-band line weights. No diffusion raster output, no broken gray lines.
  • A shape library instead of freehand geometry. LLMs reliably botch bicycles and side-view cars; the skill ships 200+ object helpers indexed by interest in `reference/catalog.md` — vehicles, trains, boats, space, farm and wild animals, buildings, playground gear, people in run/jump/point/carry/sleep poses, a wheelchair user — plus relative-proportion recipes for anything else (derived from public-domain 1910s drawing pedagogy and MIT-licensed icon skeletons — see CREDITS.md).
  • Scene kits and a composition recipe. lib/scenes.py gives proven backgrounds (meadow, street, beach, space, farm) with declared ground lines; the drawing guide's three-layer recipe (background kit → midground anchor → foreground figures at 300-380 px) is what keeps pag