gemini

Automates large-context code and plan reviews using the Gemini CLI.

Updated Mar 28, 2026
One-click install
npx skills add https://github.com/sandalsoft/agent-skills --skill gemini-sandalsoft
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gemini
Source: https://github.com/sandalsoft/agent-skills/tree/main/gemini
Command: npx skills add https://github.com/sandalsoft/agent-skills --skill gemini-sandalsoft

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Gemini-driven analysis enables comprehensive code reviews, plan reviews, and large-context processing for codebases and architectural documents.

Core Features & Use Cases

  • Code Review: Multi-file, cross-module analysis for quality, security, and maintainability.
  • Plan Review: Evaluation of technical specifications, roadmaps, and system designs.
  • Big Context Processing: Analyze entire repositories or large documentation sets that exceed typical context limits.

Quick Start

Point Gemini at the target codebase and choose the model and approval mode to start automated, large-context analysis.

Frequently Asked Questions about gemini

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate large-context code reviews for multi-file repositories?

You can automate large-context code reviews by pointing the Gemini CLI at your multi-file repository to analyze cross-module quality, security, and maintainability across over 200k tokens.

What is the best way to review architectural plans that exceed standard context limits?

Review architectural plans and large documentation sets by processing them with the Gemini CLI, which supports big context inputs exceeding 200k tokens to derive comprehensive architectural insights.

Does this code review automation support different Gemini models?

Yes, the code review automation supports model selection among gemini-3-pro-preview, gemini-3-flash, gemini-2.5-pro, gemini-2.5-flash, and gemini-2.5-flash-lite for tailored analysis.

Can I use automated approval modes for background code analysis workflows?

Yes, you can use automated approval modes including default, auto_edit, and yolo, along with interactive and background workflows to streamline continuous code analysis.

When do I need big context processing for my codebase?

You need big context processing when analyzing entire codebases or large documentation sets that exceed typical context limits, requiring over 200k tokens to derive accurate cross-module insights.