Alanamultiskillagent
Back to skills

gemini-live-api-dev

ai-tools

Covers building real-time, bidirectional streaming apps with the Gemini Live API over WebSockets. Handles audio/video/text input streams, voice activity detection, session management, ephemeral tokens

google-geminigoogle-gemini
|View all google-gemini skills →

Setup & Installation

npx skills add https://github.com/google-gemini/gemini-live-api-dev --skill gemini-live-api-dev
or paste the link and ask your coding assistant to install it
https://github.com/google-gemini/gemini-live-api-dev
View on GitHub

What This Skill Does

Covers building real-time, bidirectional streaming apps with the Gemini Live API over WebSockets. Handles audio/video/text input streams, voice activity detection, session management, ephemeral tokens, and function calling. SDKs covered are google-genai (Python) and @google/genai (JavaScript/TypeScript). Rather than piecing together WebSocket protocol details, VAD config, and session management from scattered docs, this skill gives you working patterns for the full Live API lifecycle in one place.

When to use it

  • Working with gemini live api dev functionality
  • Implementing gemini live api dev features
  • Debugging gemini live api dev related issues

Similar Skills

agent-framework-azure-ai-py

Python SDK for building persistent agents on Azure AI Foundry. Supports function tools, hosted tools (code interpreter, file search, Bing web search), MCP server integration, conversation threads, and

agents-sdk

Build stateful AI agents on Cloudflare Workers with persistent SQLite-backed state, WebSocket connections, scheduled tasks, and durable workflows. The SDK provides an Agent class with RPC methods, MCP

agents-v2-py

Deploys container-based agents to Azure AI Foundry using `ImageBasedHostedAgentDefinition` from the Azure AI Projects SDK. You provide a container image and resource specs; the SDK handles registratio

answers

Provides AI-generated answers grounded in live web search results through Brave's OpenAI-compatible chat completions endpoint. It has two modes: a fast single-search mode for quick answers, and a rese