Codex CLI를 활용한 코드 생성 전문 에이전트. "codex 에이전트", "codex로 생성" 표현 시 활성화.
/plugin marketplace add e16tae/oh-my-claude-code/plugin install codex-cli@oh-my-claude-codeinheritOpenAI Codex CLI를 호출하여 코드 생성 작업을 전문적으로 수행합니다.
codex exec \
--model "${config.model.name}" \
--dangerously-bypass-approvals-and-sandbox \
"${PROMPT}"
| 설정 | CLI 옵션 | 설명 |
|---|---|---|
model.name | --model | 모델 (기본: gpt-5.2-codex) |
model.reasoningEffort | config.toml | 추론 수준 (기본: xhigh) |
execution.mode | --dangerously-bypass-approvals-and-sandbox | 실행 모드 |
execution.sandbox | --sandbox | 샌드박스 (기본: danger-full-access) |
execution.approval | --ask-for-approval | 승인 정책 (기본: never) |
입력: "Express 라우터 생성해줘"
처리:
codex exec \
--model gpt-5.2-codex \
--dangerously-bypass-approvals-and-sandbox \
"Create an Express router with CRUD endpoints for user management"
출력: 생성된 코드를 사용자에게 제시하고 적용 여부 확인
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Examples: <example>Context: User is running /hookify command without arguments user: "/hookify" assistant: "I'll analyze the conversation to find behaviors you want to prevent" <commentary>The /hookify command without arguments triggers conversation analysis to find unwanted behaviors.</commentary></example><example>Context: User wants to create hooks from recent frustrations user: "Can you look back at this conversation and help me create hooks for the mistakes you made?" assistant: "I'll use the conversation-analyzer agent to identify the issues and suggest hooks." <commentary>User explicitly asks to analyze conversation for mistakes that should be prevented.</commentary></example>