菜单
Use when tasks benefit from Gemini CLI for code generation, review, analysis, web research, a second AI perspective, codebase architecture analysis, parallel code generation, or when the user explicitly requests Gemini operations.
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 installing, reinstalling, auditing, or preparing Brain-managed Superpowers, Spec Kit, and OpenSpec workflow arsenals under a project-local `.brain/vendor` directory so Brain can inspect and route them without exposing their raw Codex skills/prompts at the project root.
Use when the user asks to check, add, or manage macOS Calendar events.
Use when the user asks to send an email via the macOS Mail app.
Use when the user asks to send a text message or iMessage via the macOS Messages app.
Use when a task involves web search, current information, source verification, citations, or entity/person/paper/project disambiguation.
Use when the task involves `exec_command` escalation, `prefix_rule`, sandbox permissions, writable roots, or Codex platform-specific operational patterns.
| name | gemini-cli |
| description | Use when tasks benefit from Gemini CLI for code generation, review, analysis, web research, a second AI perspective, codebase architecture analysis, parallel code generation, or when the user explicitly requests Gemini operations. |
Before running Gemini, follow the shared logging and summary protocol in ../_shared/agent-cli.md.
| Use Case | Why Gemini |
|---|---|
| Current web information | google_web_search - real-time Google Search |
| Codebase architecture analysis | codebase_investigator - deep analysis tool |
| Second opinion / code review | Different AI perspective catches different bugs |
| Parallel code generation | Offload tasks while continuing other work |
When NOT to use: Simple quick tasks (overhead not worth it), interactive refinement, context already understood.
command -v geminisuperpowers:using-git-worktrees to create a fresh git worktree, then run Gemini there with write-capable flags such as --yolo when approved.the current host's user-question or approval mechanism before using --yolo or -s flags. These modes allow file writes or sandboxed execution - get explicit user approval first.-m, --model <MODEL> - Model selection-y, --yolo - Auto-approve all tool calls (enables writes)-s, --sandbox - Run in Docker isolation-o, --output-format <text|json> - Output formatfull.md, and require the prompt to write summary.md. Use -o json only when structured output is needed; preserve the raw JSON in the full log.YOLO mode does NOT prevent planning prompts. Use forceful language: "Apply now", "Start immediately", "Do this without asking for confirmation".
| Use case | Mode | Command pattern |
|---|---|---|
| Read-only analysis | read-only | gemini "..." -o json |
| Apply edits in fresh worktree only | write | gemini "..." --yolo -o json |
| Sandboxed write in fresh worktree only | sandbox | gemini "..." --yolo --sandbox -o json |
# Read-only
gemini "Review src/ for bugs" -o json
# Write mode in a fresh worktree only
gemini "Fix bug in file.py. Apply now." --yolo -o json
# Capture stdout/stderr according to ../_shared/agent-cli.md
echo "follow-up" | gemini -r latest -o jsongemini --list-sessions-m gemini-2.5-flash for lower priority tasks.gemini --version, use --debug for details.