From gemini
Quick reference for Gemini plugin commands, model aliases, effort levels, and review gate configuration. Use when you need a reminder of available options.
How this skill is triggered — by the user, by Claude, or both
Slash command
/gemini:guideThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
| Command | Purpose |
| Command | Purpose |
|---|---|
/gemini:review | Code review with full codebase context |
/gemini:challenge | Challenge-oriented design review |
/gemini:analyze | Whole-codebase analysis (1M context) |
/gemini:consult | Delegate tasks to Gemini |
/gemini:status | Check background task progress |
/gemini:result | Get finished task output |
/gemini:cancel | Stop a running task |
/gemini:setup | Check install, auth, manage review gate |
Use with --model on any command:
| Alias | Model |
|---|---|
flash, fast | gemini-3-flash-preview (default, fastest) |
pro, deep | gemini-3.1-pro-preview (most capable) |
lite | gemini-2.5-flash-lite (cheapest) |
flash-2 | gemini-2.5-flash |
pro-2 | gemini-2.5-pro |
Use with --effort on any command (maps to Gemini thinking budget):
| Level | Budget |
|---|---|
low | Light reasoning |
medium | Standard |
high | Deep reasoning |
xhigh, max | Maximum |
none | No thinking |
<number> | Token count |
/gemini:setup --enable-review-gate # Enable
/gemini:setup --disable-review-gate # Disable
/gemini:setup --gate-config model=pro # Configure
All commands route through scripts/gemini-companion.mjs. This single entry point handles:
.gemini-tasks/state.json)The stop review gate (hooks/hooks.json) uses Node.js and skips turns that didn't make code edits.
/gemini:review # Auto-detect scope
/gemini:review --model pro --effort high --base main
/gemini:analyze # Interactive menu
/gemini:consult --model flash --effort low quick fix
/gemini:challenge question the caching design
All commands work with zero arguments — they auto-detect context and present interactive menus.
npx claudepluginhub arescope/claude-plugins --plugin geminiDelegates tasks to Gemini CLI for large-context analysis like broad codebase reviews or long-document processing. Activates on explicit requests such as 'use gemini' or 'delegate to gemini'.
Runs Google's Gemini CLI for code generation, review, analysis, web research via Google Search, and codebase architecture investigation. Use for second AI opinions, real-time web data, or parallel code tasks.
Invokes Google Gemini CLI for second opinions, architectural advice, code reviews, security audits, and debugging using 1M+ context window in Claude Code workflows.