mail_search_emails
AI agents call mail_search_emails to retrieve information from Pyfastmail without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Search operations retrieve and query existing data without creating, modifying, or deleting anything. This is characteristic of Read category tools. While the Fastmail integration provides broad account access, this specific tool's function is limited to searching. Severity is low because search alone presents minimal risk—an agent cannot exfiltrate data beyond what it can already query, and no side effects occur.
From the tool's definition Tool name 'mail_search_emails' indicates querying/searching email data with no modification capability. The description is empty, but the name clearly indicates a read operation.
Attacks that exploit this kind of access
mail_search_emails. It is categorised as a Read tool in the Pyfastmail MCP Server, which means it retrieves data without modifying state.
Register the Pyfastmail MCP server in PolicyLayer and add a rule for mail_search_emails: 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 Pyfastmail. Nothing to install.
mail_search_emails 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 mail_search_emails 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 mail_search_emails. 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.
mail_search_emails is provided by the Pyfastmail MCP server (pjosols/pyfastmail-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 →