yuchi-chang
UserDrop in a picture — your cat, a doodle, a game character — and Claude analyzes it, cuts out the subject, writes it a personality, and launches it as a transparent always-on-top pet that wanders around your desktop and talks to you.
Categories
Indexed Skills (15)
desktop-pet
把任何角色或動物的圖片變成互動式桌面寵物（桌寵）。分析圖片、自動去背、產生個性化設定，並啟動一隻會在桌面走動的透明置頂寵物，支援餵食、撫摸、呼喚主人、Ctrl+點擊召喚、拖曳等互動。當使用者提供圖片並想做成桌寵時使用。Turn any character/animal image into an interactive desktop pet with auto background removal, feeding, petting, call-owner and Ctrl+click summon. Use when the user provides an image and wants a desktop pet (e.g. "make a desktop pet from this image", "把這張圖做成桌寵"), or wants to launch/manage existing pets.
clarifying-requirements
Use when starting a new feature, component, or project whose requirements are underspecified or ambiguous — before designing or writing code
durable-plans
Use when planned work will outlive the current session — persisting a plan to disk, resuming work from an on-disk plan, or landing changes a plan file describes
escalate-dont-thrash
Use when repeated attempts at the same problem keep failing without producing new evidence — before trying another variation of the last fix
exploring-approaches
Use when starting a new project or significant feature from a vague idea — before clarifying details or planning, especially when the user already has a preferred approach
verify-before-done
Use when about to claim work is complete, fixed, or passing — before committing, opening a PR, marking a task done, or relaying a subagent's success report
writing-meaningful-tests
Use when writing or modifying tests, adding mocks, or tempted to add test-only methods to production code
fail-loud
Use when writing error handling — try/catch blocks, fallback values, default parameters, retries, or any code path that handles failure
measure-before-optimizing
Use when asked to improve performance, or when about to optimize code believed to be slow
receiving-feedback
Use when receiving code review comments, reviewer suggestions, or technical corrections — before implementing any of them
root-cause-debugging
Use when encountering a bug, test failure, regression, flaky test, or unexpected behavior — before proposing or applying any fix
safe-refactoring
Use when restructuring, renaming, moving, or extracting code that already works — before making the first edit
schema-migration-safety
Use when writing database schema migrations or changing tables, columns, indexes, or constraints on a system with real data
surgical-changes
Use when editing existing code — bug fixes, small features, or any change inside a codebase you didn't just write
writing-solid-plans
Use when designing or planning a multi-step feature, refactor, or spec before implementation begins
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.