streaming-patterns

Implement real-time multimodal ADK bidi-streaming with RunConfig and LiveRequestQueue.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/RawleySM/rlm-adk --skill streaming-patterns-rawleysm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: streaming-patterns
Source: https://github.com/RawleySM/rlm-adk/tree/main/.gemini/skills/streaming-patterns
Command: npx skills add https://github.com/RawleySM/rlm-adk --skill streaming-patterns-rawleysm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-adk, google-genai, and includes scripts (resource) components.

What problem does it solve?

Enables building real-time, multimodal AI agents using ADK bidi-streaming by providing reusable patterns, templates, and workflows that handle streaming, session management, and cross-device coordination.

Core Features & Use Cases

  • Bidi-streaming patterns for voice, video, and text modalities.
  • Session management with automatic reconnection and context compression.
  • Streaming tools and multi-agent workflows for pipelined tool results and session handoffs.
  • Use cases include live voice assistants, video analysis agents, and multimodal chatbots.

Quick Start

Create a bidi-streaming RunConfig and LiveRequestQueue for a voice agent, then process streaming events.

Frequently Asked Questions about streaming-patterns

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

FAQPage Schema
How do I implement real-time multimodal streaming with ADK?

Bidi-streaming in ADK enables simultaneous bidirectional data flow for voice, video, and text by configuring a RunConfig and processing events through a LiveRequestQueue to handle real-time agent interactions.

What is a LiveRequestQueue used for in ADK streaming agents?

A LiveRequestQueue in ADK streaming manages the flow of multimodal events between clients and agents, enabling real-time processing of voice, video, and text inputs within a bidi-streaming session.

Does ADK bidi-streaming support automatic session reconnection?

Yes, ADK bidi-streaming patterns include session management capabilities with automatic reconnection and context compression to maintain live agent connectivity during real-time multimodal interactions.

Can I use ADK streaming patterns for multi-agent workflows and tool pipelining?

Yes, ADK streaming patterns support multi-agent workflows by providing pipelined tool results and session handoffs, allowing coordinated streaming across multiple agents handling real-time multimodal tasks.

What dependencies do I need to set up real-time ADK streaming sessions?

Setting up real-time ADK streaming sessions requires the google-adk and google-genai dependencies to configure RunConfig, initialize LiveRequestQueue, and manage bidi-streaming events for multimodal agents.