Gemini
Route huge repo or spec reviews through Gemini CLI when Claude needs Gemini 3 Pro-scale context and reasoning.
Overview
Gemini is an agent skill most often used in Build (also Ship, Validate) that wraps Gemini CLI so Claude Code can run Gemini 3 Pro for large-context code and plan analysis.
Install
npx skills add https://github.com/softaworks/agent-toolkit --skill geminiWhat is this skill?
- Gemini 3 Pro via Gemini CLI with very large effective context (documented up to ~1M tokens input)
- SWE-bench-oriented positioning (76.2% cited) for software engineering reasoning
- Interactive and background/automated execution modes
- Activation keywords: "use Gemini", "run Gemini", "big context", ">200k tokens"
- Use cases: whole-codebase review, architectural plans, multi-file relationship analysis
- Documented up to ~1M tokens input context via Gemini CLI
- 76.2% SWE-bench figure cited in skill overview
- 35% software-engineering improvement claim vs Gemini 2.5 Pro (per skill copy)
Adoption & trust: 3.7k installs on skills.sh; 2k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your agent cannot fit the whole codebase or spec in one pass, so architectural and cross-file reviews stay shallow or fragmented.
Who is it for?
Indie builders doing monorepo-wide reviews, roadmap reads, or >200k-token document sweeps with explicit "use Gemini" intent.
Skip if: Small single-file edits, air-gapped environments without Gemini CLI, or work that must stay entirely inside one vendor boundary.
When should I use this skill?
User asks to use or run Gemini CLI, mentions Gemini review, big context, or >200k tokens for comprehensive analysis.
What do I get? / Deliverables
Claude Code delegates a Gemini CLI run that ingests large context and returns structured analysis you can act on in the same session.
- Gemini CLI analysis output summarized for the host agent
- Cross-file or architectural findings actionable in Claude Code
- Optional background-run log when using automated mode
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Big-context analysis is most common while building and refactoring, but the same integration supports pre-ship review and architecture reads—primary shelf stays Build integrations. This skill wires an external model CLI into the agent loop rather than implementing product features.
Where it fits
Feed a full product spec and competitor notes into Gemini to sanity-check scope before committing to build.
Run Gemini across `src/` and `docs/` to map coupling before a large refactor.
Invoke Gemini for a pre-release pass over changed packages and migration scripts.
How it compares
Model-bridge integration skill—not a replacement for local `git diff` review or a dedicated static-analysis MCP server.
Common Questions / FAQ
Who is gemini for?
Solo builders and small teams using Claude Code who already use or can install Gemini CLI and need panoramic analysis across many files or huge specs.
When should I use gemini?
When you ask to run Gemini CLI, need comprehensive multi-file or whole-codebase review, analyze architectural plans, or process contexts called out as >200k tokens—during Build integration work, Validate scope review, or Ship pre-merge audits.
Is gemini safe to install?
Check the Security Audits panel on this Prism page; the skill drives an external CLI that may read repository files and send content to Google’s Gemini service per your CLI configuration.
SKILL.md
READMESKILL.md - Gemini
# Gemini Skill Harness the power of Google's Gemini 3 Pro for comprehensive code analysis, plan reviews, and big context processing. This skill enables Claude Code to leverage Gemini CLI for tasks requiring massive context windows (>200k tokens) and state-of-the-art reasoning capabilities. ## Purpose The Gemini skill provides access to Google's flagship Gemini 3 Pro model through the Gemini CLI, offering: - **Massive Context Windows**: Process up to 1M tokens of input across entire codebases - **State-of-the-Art Performance**: 76.2% on SWE-bench, 35% better at software engineering than Gemini 2.5 Pro - **Advanced Reasoning**: Exceptional at complex coding tasks, architectural analysis, and agentic workflows - **Flexible Automation**: Support for both interactive and background/automated execution modes ## When to Use This Skill Use the Gemini skill when: - User explicitly asks to run Gemini CLI or mentions Gemini - Performing comprehensive code reviews across multiple files or entire codebases - Analyzing architectural plans, technical specifications, or project roadmaps - Processing large context (>200k tokens) such as entire documentation sets - Understanding complex relationships and patterns across many files - Tasks requiring advanced reasoning on software engineering problems **Activation Keywords**: "use Gemini", "run Gemini", "Gemini review", "analyze with Gemini", "big context", ">200k tokens" ## How It Works The skill acts as an intelligent wrapper around the Gemini CLI, providing: 1. **Model Selection**: Guides users to choose the optimal Gemini model for their task 2. **Approval Mode Management**: Automatically selects the correct approval mode based on execution context 3. **Background Safety**: Prevents hung processes in non-interactive environments 4. **Structured Prompts**: Helps construct effective prompts for comprehensive analysis ### Execution Flow ``` User Request → Model Selection → Approval Mode → Command Assembly → Execution → Results ``` The skill handles all complexity around: - Preventing hung processes in background mode - Choosing between interactive and automated execution - Timeout management for safety - Error handling and recovery ## Key Features ### 1. Multiple Model Options | Model | Best For | Performance | |-------|----------|-------------| | `gemini-3-pro-preview` ⭐ | Complex reasoning, coding, agentic tasks | 76.2% SWE-bench, flagship quality | | `gemini-3-flash` | Sub-second latency, speed-critical tasks | Distilled from 3 Pro, TPU-optimized | | `gemini-2.5-pro` | Legacy all-around performance | Mature stability | | `gemini-2.5-flash` | Cost-efficient, high-volume tasks | $0.15/M tokens | | `gemini-2.5-flash-lite` | Fastest processing | Maximum speed | ### 2. Smart Approval Modes - **`default`**: Prompts for approval (interactive terminal only) - **`auto_edit`**: Auto-approves edit tools only (code reviews with suggestions) - **`yolo`**: Auto-approves all tools (required for background/automated tasks) ### 3. Background Execution Safety The skill automatically protects against hung processes by: - Enforcing `--approval-mode yolo` for non-interactive shells - Providing timeout wrappers for safety - Detecting and killing hung processes - Warning against dangerous approval mode combinations ### 4. Large Context Processing - 1M token input capacity - Multi-directory analysis support - Structured prompt guidance - Efficient codebase traversal ## Usage Examples ### Example 1: Comprehensive Code Review (Background) ```bash gemini -m gemini-3-pro-preview --approval-mode yolo \ "Perform a comprehensive code review focusing on: 1. Security vulnerabilities 2. Performance issues 3. Code quality and maintainability 4. Best practices violations" ``` **Use Case**: Automated code review in CI/CD or Claude Code background tasks ### Example 2: Architectural Plan Review ```bash gemini -m gemini-3-pro-preview --approval-mode yolo \ "Review this architectural plan