Question

Your team is deploying a conversational AI agent that connects to an MCP server exposing tools for querying and modifying network device state. Security review flags that the agent's LLM occasionally misinterprets ambiguous user requests and could trigger unintended configuration changes. Which design change best mitigates the security risk while preserving the agent's ability to answer operational questions?

Accepted Answer

Split the MCP tools so read-only query operations and write/config operations are exposed through separate servers, and grant the conversational agent access only to the read-only server for interactive Q&A

Answer

Increase the LLM's temperature setting so it produces more deterministic tool selections and avoids ambiguous configuration calls

Answer

Add a system prompt instructing the LLM to never make mistakes when calling configuration tools

Answer

Store the device credentials inside the MCP tool return payloads so the agent can validate them before every write