Which SDK should I use for extension development?

You should use the Unified Google Gen AI SDK (google-genai), as the older google-generativeai package is now deprecated for extension development.

What is a Gemini CLI Extension?

An extension is a plugin for the Gemini CLI that adds custom context (GEMINI.md), tools (MCP servers), and slash commands to the model's environment.

Can I package MCP servers in my Gemini extension?

Yes, this skill provides specific guidance on embedding MCP servers as tools within the extension architecture for use within the Gemini CLI.

Does this skill include official documentation?

Yes, it is designed to first invoke the 'gemini-cli-docs' skill, ensuring that every response is based on the most up-to-date official Google documentation.

How do I test my Gemini extension locally?

You can use the 'gemini extensions link .' command within your project directory to link your local code to the CLI, enabling hot-reloading for rapid development.

Gemini Extension Development

Name: Gemini Extension Development
Author: melodic-software

bymelodic-software

•

API開発

Builds, manages, and deploys custom extensions for the Gemini CLI ecosystem using official documentation and best practices.

概要

This skill provides an expert workflow for creating, testing, and distributing Gemini CLI Extensions, serving as a comprehensive guide for developers working within the Google Gen AI ecosystem. It streamlines the development process by managing manifest files, custom slash commands, and Model Context Protocol (MCP) server integration. By mandating a retrieval step from official docs, it ensures that all implementation patterns—from local linking to GitHub publishing—follow the latest standards for the Unified Google Gen AI SDK.

主な機能

MCP server packaging and tool integration
Standardized release workflows via Git and GitHub
Local development linking with hot-reloading support
Custom slash command definition using TOML
Step-by-step extension boilerplate creation
1 GitHub stars

ユースケース

Porting existing MCP servers into the Gemini ecosystem for enhanced model capabilities
Automating the publication and distribution of Gemini extensions to GitHub
Building a custom toolset for Gemini CLI using local scripts and context

概要

主な機能

MCP server packaging and tool integration
Standardized release workflows via Git and GitHub
Local development linking with hot-reloading support
Custom slash command definition using TOML
Step-by-step extension boilerplate creation
1 GitHub stars

ユースケース

Porting existing MCP servers into the Gemini ecosystem for enhanced model capabilities
Automating the publication and distribution of Gemini extensions to GitHub
Building a custom toolset for Gemini CLI using local scripts and context