Takes a book and distills it into a single collectible library card with cover art, author photo, and metadata. The interesting bit is the "viewfinder frame" concept: it extracts the one core insight or mental model the author contributes, explains it Feynman-style, then generates an illustrated scene showing that exact concept with a consistent character (referenced from a portrait file) in Makoto Shinkai-inspired settings. Uses WeChat Read API for Chinese book covers, Wikipedia for author photos, and Gemini for illustrations with specific aesthetic constraints (warm light, detailed objects, cozy library/workshop spaces). Output is PNG only. Invoke with book titles or phrases like "取景框卡" or "library card". Not for structural book analysis or pure text quotes, just the one-card-per-book distillation workflow.
npx -y skills add lijigang/ljg-skills --skill ljg-library --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
juliusbrussee/caveman
mattpocock/skills
obra/superpowers
forrestchang/andrej-karpathy-skills
vercel-labs/skills