菜单
This skill should be used when the user asks to "use Codex CLI", "codex exec", "get a Codex second opinion", or "run a Codex review" as a separate agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - runs review in a forked code-reviewer agent
Activate autonomous loop mode for persistent development sessions. **CRITICAL:** Always activate this AFTER completing spec and implementation plan, BEFORE beginning implementation. This is what makes autonomous builds actually autonomous—the Stop hook enforces completion criteria.
Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes
Apply systematic root cause analysis and debugging methodologies to diagnose and fix bugs, test failures, and unexpected behavior. Use when encountering production issues, investigating test failures, diagnosing performance problems, tracing error sources through call stacks, analyzing logs and stack traces, reproducing inconsistent bugs, debugging race conditions, investigating memory leaks, or applying scientific method to problem-solving before proposing fixes.
Activate orchestrator mode. You coordinate specialists, you don't implement. MAXIMIZE parallelism with task-builder. MAXIMIZE quality with code-reviewer, codex, gemini. Speed + Quality through delegation, not through doing work yourself.
Execute ONE task in parallel with other task-builders. ALWAYS SPAWN MULTIPLE for independent tasks. ORCHESTRATOR: You coordinate, you don't implement. Spawn N task-builders for N unblocked tasks. Maximum parallelism = maximum speed = maximum quality (each task gets focused attention).
| name | codex |
| description | This skill should be used when the user asks to "use Codex CLI", "codex exec", "get a Codex second opinion", or "run a Codex review" as a separate agent. |
Use Codex CLI as a parallel agent with its own context window. Prefer codex exec for headless runs and codex for the TUI.
codex exec --sandbox read-only "Review src/auth against SPEC.md. Focus on security and tests."
codex exec --full-auto --cd ../worktrees/task-3 "Implement Task 3 from IMPLEMENTATION_PLAN.md."
- to read stdin.--image, -i attaches one or more images to the first message.--model, -m overrides the configured model (e.g., gpt-5-codex).--profile, -p selects a profile from ~/.codex/config.toml.--config, -c key=value overrides config values (repeatable).--oss uses the local open source provider (requires Ollama).--cd, -C sets the working directory before execution.--skip-git-repo-check allows running outside a Git repo.--sandbox, -s: read-only | workspace-write | danger-full-access.--ask-for-approval, -a: untrusted | on-failure | on-request | never.--full-auto: sets --ask-for-approval on-request and --sandbox workspace-write.--dangerously-bypass-approvals-and-sandbox, --yolo: no approvals or sandbox.--add-dir: grant extra write roots.--search: enable web search tool usage.--json / --experimental-json: newline-delimited JSON events.--output-last-message, -o <path>: write final assistant message to a file.--output-schema <path>: validate the final response against a JSON Schema.--color <always|never|auto>: control ANSI output.codex resume [SESSION_ID] or codex resume --last [--all].codex exec resume [SESSION_ID] to continue non-interactive runs.codex fork [SESSION_ID] to branch a prior interactive session.codex exec --oss ...).