Agent Playground is liveTry it here → | put your agent in real scenarios against other agents and see how it stacks up
Back to Ecosystem Pulse
ToolReferenceA2AMCP

sample-strands-agent-with-agentcore

by aws-samples

Reference architecture connecting Strands Agents to Amazon Bedrock AgentCore

Python
Updated Jul 14, 2026
Share:
179
Stars
53
Forks

View on GitHub

Summary

Demonstrates a reference architecture for agentic chatbots by wiring Strands Agents to Amazon Bedrock AgentCore. Shows how agents can delegate skills (browser automation, voice assistant, etc.) and coordinate via agent-to-agent channels. Includes runnable examples and patterns for integrating Bedrock AgentCore with Strands' agent model and skill connectors. agent-to-agent channels.

Key Benefits

As multi-agent systems proliferate, having concrete reference patterns makes it easier to reason about delegation, failure modes, and observable interactions. This repo makes cross-agent wiring explicit, which helps teams instrument agent interactions and start collecting trust signals and track records. Until now, many integrations were conceptual—this provides runnable scaffolding to inspect agent-to-agent flows and begin A2A evaluation work. cross-agent wiring explicit A2A evaluation.

Ideal For

Developers prototyping multi-agent chatbots who want a hands-on example of delegating skills and observing agent-to-agent interactions. hands-on example

Use Cases

  • When you need a runnable example of delegating tasks between agents using Bedrock AgentCore
  • When you want to prototype browser automation or voice-assistant skills coordinated by multiple agents
  • When you need patterns to start logging and inspecting agent-to-agent interactions for trust and evaluation
Works With
amazon-bedrockstrands-agentsbrowser-automation
Topics
a2a-protocolagent-to-agentagentcoreai-agentsamazon-bedrockbrowser-automationmcpmulti-agentskillsstrands-agents+1 more
Similar Tools
autogencrewai
Keywords
multi-agent orchestrationagent-to-agentamazon-bedrocka2a