菜单
This skill should be used when the user asks to "use Gemini", "get a Gemini review", "delegate to Gemini CLI", or "run Gemini headless" for a second opinion.
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 | gemini |
| description | This skill should be used when the user asks to "use Gemini", "get a Gemini review", "delegate to Gemini CLI", or "run Gemini headless" for a second opinion. |
Run Gemini CLI in headless mode to perform tasks in this codebase.
Run with a direct prompt or stdin:
gemini --prompt "task description"
echo "Explain this code" | gemini
If the host tool supports background execution, use it for long-running calls and poll for output.
text (default) for human-readable output.json for structured output with response, stats, and optional error.stream-json for JSONL events: init, message, tool_use, tool_result, error, result.--prompt, -p - Headless prompt.--output-format - text, json, stream-json.--model, -m - Specify model (e.g., gemini-2.5-flash).--debug, -d - Enable debug mode.--include-directories - Add directories to context (comma-separated).--yolo, -y - Auto-approve all actions.--approval-mode - Set approval mode (e.g., auto_edit).| Task | Approach |
|---|---|
| Code review | Default headless prompt |
| Broad codebase question | Include paths in prompt or use --include-directories |
| Specific file analysis | Pipe file contents and include the file path in prompt |
| Structured output | --output-format json with jq -r '.response' |
| Live progress | --output-format stream-json and parse .type |
response from JSON or monitor stream-json events.gemini --prompt "Review the auth module for security issues" --output-format json | jq -r '.response'
>, >>, |.cat file.txt | gemini --prompt "Analyze this".