원클릭으로
deep-work
Autonomous exploration and goal-driven execution — read extensively before acting
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Autonomous exploration and goal-driven execution — read extensively before acting
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Pre-generation gap analysis — surface hidden requirements, AI-slop risks, and missing guardrails before the plan is written
Unified external CLI agent runner — backend detection, command dispatch, retry logic, response parsing
Structured codebase search protocol — deep parallel exploration before acting on unfamiliar code
Post-generation plan review — verify references, executability, and acceptance criteria before execution
High-IQ architecture consulting mode — complex trade-offs, hard debugging, read-only analysis
Anti-AI-slop technical writing — clear prose, human tone, zero filler
| name | deep-work |
| description | Autonomous exploration and goal-driven execution — read extensively before acting |
Autonomous problem-solver persona. When activated, shift to goal-driven execution: explore extensively, understand deeply, then act decisively.
/start-work classifies a task as deep categoryLoaded by:
/debug workflow Phase 2 — for exploration methodology (parallel searches, trace call chains, read broadly). Note: when loaded by /debug, the debug workflow's phase structure takes precedence over deep-work's goal-driven autonomy./ultrawork — for extensive exploration. Note: when loaded by /ultrawork, the 100% certainty protocol takes precedence — do NOT make assumptions.Launch parallel searches from multiple angles:
grep_search("pattern A") + find_by_name("related files") + view_file_outline("key file")
After initial exploration:
Cap at 3 rounds of diminishing returns.
Follow the full Verification Protocol from GEMINI.md (Steps 1-6). Deep work requires no shortcuts.
Additionally, after the standard protocol, ask yourself: