Gemini API Development
Official
API開発
Builds high-performance AI applications using the latest Gemini and Gemma models with modern SDKs and multimodal capabilities.
The Gemini API Development skill provides Claude with the most current specifications and implementation patterns for Google's Gemini ecosystem. It ensures developers use up-to-date models like Gemini 3.5 and 3.1 while avoiding deprecated legacy SDKs and older model versions. This skill facilitates complex tasks including multimodal processing of text, images, and video, as well as advanced features like function calling and structured outputs. By integrating with MCP documentation tools, it provides a real-time source of truth for Python, JavaScript, Go, and Java development.
主な機能
01Precise model selection for reasoning, speed, or cost-efficiency
02Support for Gemini 3.5 Flash, 3.1 Pro, and Gemma 4 model families
03Implementation patterns for multimodal content and function calling
04Modern SDK guidance for Python, JS/TS, Go, and Java
05Real-time documentation access via MCP and structured LLMS indexes
063,794 GitHub stars
ユースケース
01Developing multimodal applications that process video, audio, and images
02Implementing structured data extraction and automated tool-calling workflows
03Migrating legacy Gemini integrations to the modern 'google-genai' SDKs