An enterprise multi-agent platform must let planning agents discover and call external tools from multiple teams and must also allow a partner’s independently hosted agent to collaborate on shared cases. Which protocol selection best matches these integration needs?
Select an answer to reveal the explanation.
Short Explanation
Correct answer: D. AI-500 expects familiarity with MCP for tools/knowledge and A2A for integrating existing or external agents. Ad-hoc SSH JSON (A), FTP drops (B), and SMTP control planes (C) lack modern agent contracts, discovery, and security patterns. Specify protocols during architecture so identity, schemas, and observability can be designed around them.
Full Explanation
Architects must specify appropriate protocols for communication among agents and between agents and other components. MCP standardizes how hosts connect to tool and data servers (including Azure Functions, Logic Apps, API Management implementations). A2A addresses agent-to-agent interoperability when systems are not a single monolith. Option A creates brittle, non-auditable integrations. Options B and C are unsuitable real-time control planes for tool invocation and multi-agent collaboration. In design docs, map which interactions are tool calls (MCP) versus peer agent collaboration (A2A), including authentication methods and permission boundaries on each channel.