Your team must expose an internal pricing calculator to multiple agents as a governed tool. Developers prefer a lightweight serverless implementation that can be reached through Model Context Protocol clients. Which implementation approach best fits Azure multi-agent tool ecosystems?
Select an answer to reveal the explanation.
Short Explanation
Tools should be real services, not prompt folklore. Host the pricing calculator as an MCP server on Azure Functions (or comparable Azure host), authenticate clients, and let agents call it through MCP. Rate changes update one service. That is option B. Prompt-hardcoding (A) drifts. Spreadsheet-by-email (C) is not a tool interface. RDP UI clicking (D) is brittle and insecure.
Full Explanation
AI-500 expects designing and building MCP servers and clients, including Azure Functions, Logic Apps, and API Management patterns. Option B is the idiomatic governed tool path for multi-agent consumption.
Option A duplicates logic and breaks single-source-of-truth pricing.
Option C is not machine-actionable or reliable.
Option D is an anti-pattern for automation reliability and security.
Also apply tool scopes, auth, and result validation around the MCP tool.
Exam tip: MCP is how agents discover and invoke tools/resources consistently across a multi-agent estate.