Fetch unread emails from Gmail
AI agents call get_unread_emails to retrieve information from Gmail Assistant MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves email messages (a Read operation). Severity is medium rather than low because unread emails often contain sensitive personal, financial, or business information that an AI agent could inadvertently expose, leak, or process insecurely. However, it causes no data mutation or destructive side effects, so it does not exceed the Read category.
From the tool's definition Tool description: 'Fetch unread emails from Gmail' — retrieves email data without modification or deletion.
Attacks that exploit this kind of access
Fetch unread emails from Gmail. It is categorised as a Read tool in the Gmail Assistant MCP MCP Server, which means it retrieves data without modifying state.
Register the Gmail Assistant MCP server in PolicyLayer and add a rule for get_unread_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 Gmail Assistant MCP. Nothing to install.
get_unread_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 get_unread_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 get_unread_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.
get_unread_emails is provided by the Gmail Assistant MCP server (thejoefriel/gmail-assistant-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 →