AI agents call searchMessages to retrieve information from Outlook without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
searchMessages retrieves or queries email messages without modifying them. This is a classic Read operation with no side effects. While the tool description is absent, the name unambiguously indicates a search function, which is non-destructive data retrieval. Severity is low because searching email has minimal blast radius—it cannot delete, modify, or move messages, only expose existing data.
From the tool's definition Tool name 'searchMessages' indicates a query operation on email data. The sibling tools on the Outlook MCP server include retrieval operations (getMessage, getRecentMessages, listAccounts, listFolders, listCalendars) and modification operations (createEvent,…
Attacks that exploit this kind of access
searchMessages. It is categorised as a Read tool in the Outlook MCP Server, which means it retrieves data without modifying state.
Register the Outlook MCP server in PolicyLayer and add a rule for searchMessages: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Outlook. Nothing to install.
searchMessages is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the searchMessages rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for searchMessages. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
searchMessages is provided by the Outlook MCP server (lihaokun/outlook-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →