Back to directory

Multi-agent Orchestration Chat on AgentCore

An AWS sample multi-agent chat platform, built on Amazon Bedrock AgentCore and serverless TypeScript services, letting a team create, customize and share AI agents, with preset agents, event-driven triggers and extensible tools.

Multi-agent Orchestration Chat on AgentCore

This AWS sample is a multi-agent orchestration chat platform built on Amazon Bedrock AgentCore, letting a team create and customize AI agents and share them across the organization. Preset agents, such as a software developer, a data analyst and a physicist, are available immediately, alongside custom agent creation, organization-wide agent sharing and discovery, scheduled or event-driven automatic triggering, extensible tools for command execution, web search, image generation and external service integration, and built-in cloud storage for documents and resources.

The architecture is fully serverless: a React single-page frontend served through CloudFront and S3 authenticates through Amazon Cognito, and requests flow through a Lambda and API Gateway backend to an AgentCore Runtime that orchestrates agent execution through the AgentCore Gateway, using the Strands Agents SDK for TypeScript with both short-term session memory and long-term persistent memory, DynamoDB and S3 for storage, and AppSync Events over WebSocket for real-time updates, with EventBridge Scheduler and Rules driving scheduled and event automation.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.