A2A Research Digest — 2026/03/13: Survey of Agent Interoperability Protocols
Paper 1: Survey of Agent Interoperability Protocols
Title: A survey of agent interoperability protocols: Model Context Protocol (MCP), Agent Communication Protocol (ACP), Agent-to-Agent Protocol (A2A), and Agent Network Protocol (ANP)
Authors: Abul Ehtesham, Aditi Singh, Gaurav Kumar Gupta, Saket Kumar
Date: May 4, 2025 (v2: May 23, 2025)
arXiv: 2505.02279
Core Contributions
- First Comprehensive Comparison - Analyzes four major agent interoperability protocols: MCP, ACP, A2A, and ANP
- Discovery Mechanisms - Detailed analysis of how agents discover each other
- Communication Patterns - Synchronous vs asynchronous, push vs pull models
- Security Models - Authentication, authorization, and trust frameworks
- Phased Adoption Roadmap - MCP → ACP → A2A → ANP
Technical Summary
| Protocol | Type | Key Feature |
|---|---|---|
| MCP | Client-Server | JSON-RPC for tool invocation |
| ACP | RESTful HTTP | MIME multipart messages, session management |
| A2A | Peer-to-Peer | Agent Cards for capability-based discovery |
| ANP | Decentralized | W3C DID, JSON-LD for open networks |
Paper 2: Building Secure A2A Applications
Title: Building A Secure Agentic AI Application Leveraging A2A Protocol
Authors: Idan Habler, Ken Huang, Vineeth Sai Narajala, Prashant Kulkarni
Date: April 23, 2025 (v2: May 2, 2025)
arXiv: 2504.16902
Core Contributions
- First Security Analysis - Comprehensive security analysis of A2A protocol
- Threat Modeling - Uses MAESTRO framework for AI risk assessment
- Agent Card Security - Management and verification of agent identity
- Task Execution Integrity - Ensuring reliable task completion
- A2A + MCP Synergy - How protocols work together for enhanced security
Security Threats Identified
- Agent Impersonation - Fake agents in discovery
- Task Hijacking - Intercepting and modifying tasks
- Message Tampering - Altering communication content
Recommended Best Practices
- Zero Trust Architecture
- Input Validation
- Audit Logging
- OAuth 2.0 + JWS for authentication
- Agent Card Signature Verification
Analysis & Insights
These two papers represent the foundational academic research on A2A protocol:
- Protocol Comparison Paper provides the architectural overview
- Security Paper addresses critical real-world deployment concerns
Key Takeaway: The combination of MCP (agent-to-tool) and A2A (agent-to-agent) creates a comprehensive stack for building agentic applications. Security must be designed from the ground up.
Tags: a2a, research, papers, security, protocols