AI News

Trending story

AWS releases cross-organizational multi-agent reference architecture based on A2A

1 reports1 story updatesUpdated Nov 12 23:42

Understand this story

AI overview

On November 12, 2026, AWS released a cross-organizational multi-agent system reference architecture based on the A2A protocol. This architecture enables a LangChain-based orchestration agent within an enterprise account and a Strands Agents-based compliance agent within a SaaS account to collaborate across organizations, with neither party depending on the other's framework or SDK, nor sharing keys. The solution uses Amazon Cognito to implement bidirectional OAuth authentication, and the SaaS agent can call back to the enterprise context agent to obtain account metadata; the entire system is deployed through three Jupyter notebooks and has completed end-to-end testing. There are currently no subsequent developments.

AI-generated from public reports

Latest development · Report

AWS Publishes Tutorial on Deploying Cross-Organization Multi-Agent Systems Based on A2A

Developments

1 confirmed developments

  1. AWS releases cross-organizational multi-agent reference architecture based on A2A

    Source:AWS Machine Learning

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 12

  1. AWS Machine LearningFirst-party

    AWS Publishes Tutorial on Deploying Cross-Organization Multi-Agent Systems Based on A2A

    AWS has released a reference architecture based on the A2A protocol that enables an orchestration agent built on LangChain within an enterprise account to collaborate across organizations with a compliance agent built on Strands Agents within a SaaS account, without either party depending on the other's framework or SDK, and without sharing keys. The solution uses Amazon Cognito to implement bidirectional OAuth authentication; the SaaS agent can call back to the enterprise context agent to obtain account metadata, and the entire system is deployed and end-to-end tested through three Jupyter notebooks.

Related stories

  1. Related storiesViquar Khan demonstrates building autonomous AI agents with Spring AI and Bedrock
  2. Related storiesEnglish translation is being prepared.
  3. Related storiesThe DeepSeek Harness team shares its Super-agent design exploration
  4. Related storiesDigital Life Kazik shares experience building a self-hosted Agent automation system