kwakseongjae

One-prompt autonomous product design and implementation. Use automatically for broad greenfield UI requests such as 'from scratch', '새 제품/화면을 알아서 만들어줘', or requests that delegate DESIGN.md creation. It decides whether to reuse, establish, refresh, or skip a…

One-prompt autonomous product design and implementation. Use automatically for broad greenfield UI requests such as 'from scratch', '새 제품/화면을 알아서 만들어줘', or requests that delegate DESIGN.md creation. It decides whether to reuse, establish, refresh, or skip a…

사용자와 단계별로 검토하는 guided design harness. Discovery→Wireframe→Components→Microcopy→Validation을 omd-master가 실행하고 journey/system/validation의 필수 체크포인트에서 멈춘다. '/omd-harness', '체크포인트마다 검토', '나와 단계별로 디자인', 'guided design' 요청에 사용. 질문 없이 원샷으로 새 제품을 자율 구축하는 요청은…

사용자와 단계별로 검토하는 guided design harness. Discovery→Wireframe→Components→Microcopy→Validation을 omd-master가 실행하고 journey/system/validation의 필수 체크포인트에서 멈춘다. '/omd-harness', '체크포인트마다 검토', '나와 단계별로 디자인', 'guided design' 요청에 사용. 질문 없이 원샷으로 새 제품을 자율 구축하는 요청은…

Reference batch writer compatibility gate during the DESIGN.md Core v2 catalog cutover.

Legacy catalog component writer safety gate. Use when component harvesting is requested, but stop before editing until the Core v2 graph-backed catalog writer and readers are available.

Archived DESIGN.md impact experiment compatibility gate. Use when an old Lab #01 run is requested, but do not generate new legacy 9/12-section documents; direct new experiments to the Core v2 benchmark track.

Legacy deep-reference writer safety gate. Use for old brand migration requests only to prevent retired section/frontmatter writes until the Core v2 catalog migration pipeline is authoritative.

Stops Claude from stacking guesses on top of unknowns. When uncertainty appears in code, configs, APIs, or error messages, this skill forces a brief pause to verify — by reading source, running a small check, or asking the user — instead of fabricating a…

Blocks speculative code, options, abstractions, and flags built for hypothetical future needs (YAGNI). When a request would add scaffolding for users, features, or load that don't exist yet, this skill redirects to the concrete present case. 오지 않은 미래 요구를 위해…

Forces the WHY before the HOW when code is being written, edited, reviewed, refactored, or generated. Surfaces the intent behind a change in one sentence so the result is purposeful rather than merely working, and prefers not writing code at all when intent…

Enforces minimal-code-for-same-output discipline. Before letting new functions, files, modules, abstractions, or dependencies grow the codebase, this skill checks whether the same outcome can be achieved by extending what already exists, and gives a…

Leaves the WHY beside non-obvious decisions — in commit messages, PR descriptions, and short comments next to surprising code. Never writes ceremonial comments that restate the obvious WHAT; only surfaces the non-obvious reasoning that would otherwise be lost…

Treats failure as a precondition for risky work. Before deploys, data deletions, migrations, force pushes, dependency upgrades, or other hard-to-undo actions, this skill builds the retreat path first — dry-run, backup, staged rollout, explicit rollback…

Surveys what already exists before writing new code, and refuses to delete code whose purpose is unknown. When a request would replace, remove, or duplicate existing functionality, this skill greps the codebase, reads git history for buried reasons, and…

Forces reproduction before any fix. When something doesn't work, this skill stops the symptom→guess→patch loop and first reproduces the bug locally on the same branch, data, and inputs the user saw — because an unreproduced bug is an unsolved one. 버그·이상 동작…