google-gemini-api

Integrate Google Gemini API with the GenAI SDK for multimodal AI workflows.

18|4|Updated Jan 15, 2026

Ready to run

npx skills add https://github.com/ataschz/tanstack-start-mastra-example --skill google-gemini-api-ataschz

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: google-gemini-api
Source: https://github.com/ataschz/tanstack-start-mastra-example/tree/main/.agents/skills/google-gemini-api
Command: npx skills add https://github.com/ataschz/tanstack-start-mastra-example --skill google-gemini-api-ataschz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a ready-to-use blueprint for integrating Google Gemini API with the @google/genai SDK. It covers text generation, multimodal inputs (images, video, audio, PDFs), function calling, streaming, and thinking mode, enabling robust AI-powered apps while avoiding deprecated SDK pitfalls.

Core Features & Use Cases

Text generation with Gemini models (gemini-2.5-pro/flash) using the GenAI SDK
Multimodal inputs: image, video, audio, PDFs in a single prompt
Tooling: function calling to connect external APIs
Streaming responses and multi-turn conversations
thinking mode configuration for advanced reasoning
Guidance for caching, grounding, and code execution workflows

Quick Start

Install dependencies: npm install @google/[email protected]
Initialize the Gemini client and run a basic generation example
Explore code samples in the references/templates for onboarding

Frequently Asked Questions about google-gemini-api

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I integrate Google Gemini for multimodal AI workflows using the GenAI SDK?▼

Integrate Google Gemini for multimodal AI workflows by using the @google/genai SDK to process text, images, video, audio, and PDFs. This approach provides a blueprint for building chatbots and assistants while enforcing correct SDK usage and explicit model versions.

Can I use function calling and streaming responses with Gemini models in my application?▼

Yes, you can use function calling and streaming responses with Gemini models. The GenAI SDK integration supports connecting external APIs via function calling and enables streaming responses for multi-turn conversations within your AI-powered applications.

Does the Gemini API support image, video, and audio inputs in a single prompt?▼

The Gemini API supports image, video, audio, and PDF inputs in a single prompt. This multimodal capability allows developers to build complex workflows that process diverse data types simultaneously using Gemini models like gemini-2.5-pro or flash.

How do I configure thinking mode and context size limits for Gemini models?▼

Configure thinking mode and context size limits by enforcing explicit model versions and size constraints within your GenAI SDK code. This ensures advanced reasoning capabilities are properly managed while avoiding deprecated SDK pitfalls during generation.

What is the best way to avoid deprecated SDK pitfalls when building Gemini AI apps?▼

The best way to avoid deprecated SDK pitfalls when building Gemini AI apps is to use the @google/genai SDK version 1.27.0 or later. This enforces correct SDK usage patterns, including proper handling of caching, grounding, and code execution workflows.

Are there limitations when using caching and grounding patterns with Google Gemini?▼

Limitations when using caching and grounding patterns with Google Gemini relate to context size limits and explicit model version requirements. Developers must enforce correct SDK usage to manage these constraints properly within their multimodal AI workflows.