Loading...
Loading...
Found 80 Skills
AI 에이전트와 협업하는 에이전틱 개발의 범용 원칙. 분해정복, 컨텍스트 관리, 추상화 수준 선택, 자동화 철학을 정의. 모든 AI 코딩 도구에 적용 가능.
Use only then explicitly asked.
Design stress-test engine. Does NOT generate designs — it pressure-tests YOUR existing draft against real-world edge cases, code constraints, and failure paths. One sharp question at a time, drilling down the decision tree, using code facts to stress-test assumptions, then outputs a tight Facet Brief. NOT for: executing clear tasks, factual Q&A, or pure brainstorming (go brainstorm first, then come back with a direction). Triggers: "facet / pressure-test this plan / stress-test the design".
Runs the full validator workflow after coding tasks for requests such as "run the validator", "run final verification", "validate before commit", or "run validation". Executes checks and reviews before commit, push, or PR creation.
Verify review or audit findings and fix confirmed issues
Execute a written implementation plan task-by-task with inline verification. Stop and ask on blockers. No forced sub-skill chains. Use when a plan exists and you're ready to implement.
Run the full spec-driven workflow automatically. Proposes, implements, verifies, reviews, and archives a change with one mandatory proposal checkpoint plus any extra confirmations required by blocking conditions.
Identify, analyze, and document malware distribution repositories masquerading as legitimate software
Use when about to say "done", "fixed", "passing", or "shipped" - or when reporting the outcome of any change. Encodes verification as the definition of done: run the proof, report faithfully, distrust green suites, own failing gates. Use before every completion claim, even when the change was small and obviously correct, which is when this is skipped.
Fixes blocking IO, unnecessary allocations, sync-in-async with keep/discard verification
unity-cli 共通ルール。全スキルの前提条件。
Finish an open PR by batching review feedback, applying valid fixes, running verification, pushing once per batch, and monitoring CI with bounded waits. Use when the user asks to update code until PR review comments are handled, review-all has no valid blocking findings, and required checks pass.