Can I use this for Gemini function calling?

Yes, it includes detailed examples and code snippets for implementing native function/tool calling and generating validated JSON outputs with schema enforcement.

Which Gemini models are covered?

The skill provides a capability reference and prompting strategies for the Gemini 3 and Gemini 2.5 model families, including both Flash and Pro variants.

Does this skill support Gemini's multimodal features?

Yes, it includes specific implementation patterns for prompting with images, audio, and video alongside traditional text inputs.

What is the Gemini Prompt Engineering skill?

It is a specialized guide for Claude Code that provides best practices, templates, and architectural patterns for optimizing interactions with Google's Gemini models.

How does it handle Gemini's long context window?

The skill provides strategic guidance for effectively utilizing Gemini's 1M+ token context to synthesize themes across massive document sets or large-scale codebases.

Gemini Prompt Engineering

Name: Gemini Prompt Engineering
Author: duyet

byduyet

•

데이터 과학 및 ML

Optimizes AI workflows by providing expert guidance and implementation patterns for Google Gemini models.

소개

This skill empowers Claude to assist developers in crafting high-performance prompts specifically for the Google Gemini ecosystem. It covers advanced techniques like system instructions, multimodal input handling, and utilizing ultra-long context windows up to one million tokens. Whether you are building complex agents with ReAct patterns or need structured JSON outputs, this skill provides the necessary templates and best practices to leverage Gemini's unique reasoning and multimodal capabilities effectively.

주요 기능

Structured output configuration using JSON Schema and function calling
Ultra-long context strategies for managing up to 1M+ tokens
Multimodal prompting patterns for images, audio, and video inputs
Advanced System Instruction templates for precise behavior control
2 GitHub stars
Complex reasoning frameworks including Chain-of-Thought and ReAct

사용 사례

Migrating or optimizing existing prompts for the Gemini 3 and 2.5 model families
Developing AI agents that require native multimodal reasoning and tool use
Analyzing massive codebases or document sets using industry-leading context windows

소개

주요 기능

Structured output configuration using JSON Schema and function calling
Ultra-long context strategies for managing up to 1M+ tokens
Multimodal prompting patterns for images, audio, and video inputs
Advanced System Instruction templates for precise behavior control
2 GitHub stars
Complex reasoning frameworks including Chain-of-Thought and ReAct

사용 사례

Migrating or optimizing existing prompts for the Gemini 3 and 2.5 model families
Developing AI agents that require native multimodal reasoning and tool use
Analyzing massive codebases or document sets using industry-leading context windows