Does this skill provide guidance on the latest Google SDKs?

Yes, it specifically promotes the use of the modern @google/genai SDK and provides implementation examples, replacing the deprecated @google/generative-ai library.

How does Gemini 3+ handle reasoning configurations?

Gemini 3+ models use a 'thinkingLevel' parameter (low, medium, high) within the generationConfig, rather than the legacy 'thinkingBudget' found in older versions.

Can I use Google Search and custom functions simultaneously?

Yes, but they must be implemented as separate tool objects within the tools array to avoid 400 errors, a specific pattern detailed in this skill.

Which Gemini models are covered in this reference?

It covers the latest Gemini 3.x models, including 3.1 Pro Preview and 3.0 Flash, while identifying legacy models like 1.5 and 2.0 as deprecated.

Gemini API Reference

Name: Gemini API Reference
Author: MeechYourGoals

byMeechYourGoals

•

数据科学与机器学习

Provides current Google Gemini API specifications, implementation patterns, and SDK usage for high-performance AI application development.

This skill serves as a specialized knowledge base for developers working with the Google Gemini API, specifically within the context of the Chravel ecosystem. It provides up-to-date technical specifications for Gemini 3.1 and 3.0 models, including advanced features such as thinking configurations (thinkingLevel), incremental streaming for function call arguments, and combined grounding techniques. By centralizing the latest @google/genai SDK practices and Chravel-specific integration points, it enables developers to build sophisticated AI features like real-time travel concierges, smart data parsing, and multimodal interactions with precision and efficiency.

主要功能

01Guidance on combined grounding with Google Search and Maps API

02Detailed specifications for Gemini 3.1 Pro and 3.0 Flash models

03Implementation patterns for the modern @google/genai SDK

04Advanced tool-use configurations including thinkingLevel and streaming arguments

052 GitHub stars

06Direct reference links to official Gemini API documentation and discovery paths

使用场景

01Implementing structured data extraction from documents using Gemini models

02Optimizing LLM performance with incremental function call streaming

03Building AI-driven concierge services with real-time reasoning and tool use

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add meechyourgoals/chravel chravel-gemini-api-ref

For use in Claude.ai and ChatGPT

主要功能

01Guidance on combined grounding with Google Search and Maps API

02Detailed specifications for Gemini 3.1 Pro and 3.0 Flash models

03Implementation patterns for the modern @google/genai SDK

04Advanced tool-use configurations including thinkingLevel and streaming arguments

052 GitHub stars

06Direct reference links to official Gemini API documentation and discovery paths

使用场景

01Implementing structured data extraction from documents using Gemini models

02Optimizing LLM performance with incremental function call streaming

03Building AI-driven concierge services with real-time reasoning and tool use