How does it handle very large codebases?

Gemini supports a context window of over 1 million tokens, allowing it to process and understand entire repositories or extremely long log files that would typically exceed Claude's native limit.

What model does the Gemini skill use?

The skill specifically utilizes the gemini-3-pro-preview model to ensure the highest level of reasoning, multimodal support, and the largest available context window.

Can I use this skill to analyze non-text files like images or PDFs?

Yes, it is designed for multimodal tasks. You can use it to analyze images (.png, .jpg), videos (.mp4), audio files, and PDFs by using triggers like 'analyze image' or 'read PDF'.

Can Gemini provide up-to-date information from the web?

Yes, Gemini includes real-time grounding capabilities. When asked for 'latest info' or 'current news', it can search the web and provide answers with linked sources.

Does this skill require manual confirmation for every action?

No, it is configured with a --yolo flag by default to bypass permission prompts, enabling smooth, non-interactive execution within your terminal environment.

Gemini AI Orchestrator

Name: Gemini AI Orchestrator
Author: e16tae

bye16tae

0•

Data Science & ML

Integrates Google's Gemini-3-pro-preview model into Claude Code for multimodal analysis, massive context processing, and real-time grounding.

The Gemini skill empowers Claude Code with the advanced capabilities of Google's Gemini-3-pro-preview model via a non-interactive CLI interface. It excels at tasks that exceed standard LLM boundaries, such as processing images, PDFs, videos, and audio files, while seamlessly managing massive repositories using its 1-million-token context window. Whether you need deep architectural audits, real-time web searches for the latest documentation, or comprehensive PR reviews of large code changes, this skill provides high-depth reasoning and automated execution to streamline complex development workflows.

Key Features

01Massive 1M+ token context window for analyzing entire codebases at once

020 GitHub stars

03High-depth reasoning optimized for security audits and architectural reviews

04Multimodal support for analyzing images, videos, audio, and PDF documents

05Real-time web grounding for up-to-date information and source citations

06Automated CLI execution with YOLO mode for seamless non-interactive workflows

Use Cases

01Reviewing large Pull Requests by ingesting the entire context of all changed files simultaneously

02Analyzing complex UI screenshots or architecture diagrams to generate implementation plans

03Performing full-repository security audits to identify logic flaws and vulnerabilities

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add e16tae/oh-my-claude-code gemini

For use in Claude.ai and ChatGPT

Download Skill