The community node brings persistent memory, real tools, and multi-provider model support to n8n workflows.
AWS announced that Amazon Bedrock AgentCore harness is generally available. A new open-source community node, @aws/n8n-nodes-agentcore, brings the harness into the n8n visual editor, letting users run production agents with persistent memory, real tools, and model choice without writing infrastructure.
Amazon Bedrock AgentCore is the managed platform that runs the agent orchestration loop, manages the context window, keeps state across turns, and calls tools. The new node is MIT-licensed and works with Amazon Bedrock, OpenAI, Google Gemini, and LiteLLM-supported providers. The node also lets you switch providers between turns of the same conversation.
For n8n users, the node removes the need to build agent scaffolding themselves. They get persistent memory, real tools such as a code interpreter, and the ability to scope memory to individual users. Enterprises can also run agents privately in their own VPC using the same AWS credential patterns.
The node is open source under MIT and AgentCore harness runs on Strands Agents. Teams can export harness configurations to Strands code when needed, and paste an existing harness ARN into n8n to invoke external agents. Watch for reusable agent configurations to spread through the n8n community.
What matters
- AWS made AgentCore harness generally available and released a community node for n8n.
- Builders get persistent memory, real tools, and model choice without writing agent infrastructure.
- Watch for the node to support VPC deployment and switching providers mid-conversation.
Why it matters
Watch for the node to support VPC deployment and switching providers mid-conversation.
This GenAI News article was prepared in original wording using reporting and materials published by AWS Machine Learning Blog. Source reference: https://aws.amazon.com/blogs/machine-learning/run-production-ai-agents-in-n8n-with-amazon-bedrock-agentcore-harness/.
Drafted by the GenAI News review pipeline.