How do I prevent Gemini from hanging in background tasks?

Always use the '--approval-mode yolo' flag when running Gemini via Claude Code tool calls. The 'default' mode requires an interactive terminal and will hang indefinitely in automated environments.

Does this skill support multi-directory analysis?

Yes, you can use the '--include-directories' flag to include multiple paths in the workspace for comprehensive cross-project analysis.

What version of the Gemini CLI is required?

This skill requires Gemini CLI v0.16.0 or later to ensure full support for Gemini 3 models and the specialized approval modes.

When should I use the Gemini skill over standard Claude tools?

Use the Gemini skill for tasks requiring more than 200k tokens of context, such as analyzing an entire repository, or when you need Gemini 3 Pro's specific reasoning for complex software engineering tasks.

Which Gemini model is best for coding?

The 'gemini-3-pro-preview' is the flagship model and is recommended for complex reasoning, agentic tasks, and software engineering due to its high accuracy and large output window.

Gemini CLI Integration

Name: Gemini CLI Integration
Author: Microck

byMicrock

•

보안 및 테스팅

Integrates Gemini 3 Pro into Claude Code for massive context processing and high-fidelity automated code reviews.

The Gemini skill empowers developers to leverage Google's state-of-the-art Gemini 3 models directly within Claude Code, specifically optimized for tasks requiring massive context windows up to 1M tokens. It is designed for comprehensive codebase analysis, architectural plan reviews, and deep-dive security audits that exceed the standard context limits of individual LLM turns. With specialized handling for non-interactive background modes and a focus on 'vibe coding' and agentic workflows, this skill provides a robust bridge to Gemini's top-tier software engineering reasoning capabilities.

주요 기능

01Flexible Model Selection: Support for Gemini 3 Pro, Flash, and legacy 2.5 models depending on speed and cost needs.

02Automated Background Mode: Optimized 'yolo' approval modes specifically designed for non-interactive tool calls.

03Comprehensive Code Reviews: Multi-file analysis for security vulnerabilities, performance bottlenecks, and technical debt.

04Gemini 3 Pro Reasoning: Access state-of-the-art coding and reasoning performance (76.2% on SWE-bench).

05Large Context Window: Process entire codebases and documentation sets with up to 1M token support.

062 GitHub stars

사용 사례

01Conducting full-repository security and maintainability audits across hundreds of files.

02Deep analysis of massive documentation sets or legacy codebases to identify refactoring opportunities.

03Reviewing complex architectural plans and technical specifications for scalability and integration gaps.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microck/opendots-microck gemini

For use in Claude.ai and ChatGPT