Agent-to-Agent Communication: Connecting Foundry and LangGraph Agents via A2A

2026-07-01 · Source: Microsoft Foundry Blog articles · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure, Robotics & Autonomous Systems · Depth: Intermediate, short

Summary

A tutorial published on Jul 01, 2026, demonstrates Agent-to-Agent (A2A) communication, specifically integrating a Microsoft Foundry agent with a LangGraph agent hosted on Azure Container Apps. This setup addresses the challenge of inter-framework agent discovery and collaboration in multi-agent workflows. The LangGraph agent, built with langgraph==1.2.6 and openai azure-identity, functions as an A2A server, summarizing user input using the gpt-4.1-mini model with API version 2024-12-01-preview. It is containerized using python:3.12-slim and exposed via port 8000 with fastapi==0.138.0 and a2a-sdk[http-server]>=0.3.0,<0.4. The Foundry agent acts as the A2A client, configured through the Microsoft Foundry portal to call the LangGraph agent's endpoint for summarization tasks. The process involves deploying the LangGraph agent to Azure Container Apps and then connecting it as a custom A2A tool within Foundry.

Key takeaway

For AI Engineers building multi-agent systems, if you are integrating agents from different frameworks, you should adopt the A2A protocol to enable seamless communication. This allows your Microsoft Foundry agents to leverage specialized LangGraph agents, like a gpt-4.1-mini summarizer, hosted on Azure Container Apps. Consider implementing authentication and streaming responses for production readiness, and explore connecting multiple A2A agents to build complex workflows.

Key insights

Connecting diverse AI agents across frameworks is achievable via a standardized A2A protocol and server-client architecture.

Principles

• Multi-agent systems require robust inter-framework communication.
• Standardized protocols facilitate agent discovery and interaction.
• Agent metadata (Agent Card) enables client-side tool integration.

Method

Develop a LangGraph agent (e.g., a summarizer using gpt-4.1-mini). Expose the agent as an A2A server using a2a-sdk and Starlette. Containerize and deploy the server to Azure Container Apps. Configure a Microsoft Foundry agent to use the deployed A2A endpoint as a custom tool.

In practice

• Host LangGraph agents on Azure Container Apps for A2A.
• Use a2a-sdk for exposing agent capabilities.
• Integrate external agents as custom tools in Microsoft Foundry.

Topics

• Agent-to-Agent Communication
• Microsoft Foundry
• LangGraph
• Azure Container Apps
• Multi-Agent Systems
• AI Agent Integration
• GPT-4.1-mini

Code references

• SonakshiA/agent2agent-foundry

Best for: AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• Use A2A to connect agents across different frameworks and teams — A2A standardizes agent communication, enabling collaboration and reuse across diverse frameworks and teams.
• How Microsoft Used Its Own Tech To Lead The Gartner® Magic Quadrant™ for AI Application Development — Microsoft Foundry and its Agent Framework streamline multi-agent AI application development from ideation to production.
• Build highly scalable serverless LangGraph multi-agent systems in AWS with Amazon Bedrock AgentCore — LangGraph, AWS serverless, and Amazon Bedrock AgentCore enable scalable, observable, and stateful multi-agent generative AI systems.
• Bring Your Own Model to Foundry Agent Service Is Now Generally Available — Foundry Agent Service's BYOM feature allows enterprises to integrate custom or third-party AI models via their own gateways.
• Article: Architecting Agentic MLOps: A Layered Protocol Strategy with A2A and MCP — Layering A2A for communication and MCP for capabilities enables robust, extensible multi-agent MLOps workflows.
• Retrofit, don’t rebuild: Agentic overlays for transforming legacy enterprise services — Agentic overlays enable legacy REST services to participate in A2A communication without costly refactoring.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Microsoft Foundry Blog articles.