Fetch the full content of an email by its UID (from search_emails results).
AI agents call read_email to retrieve information from Claude Gmail without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves email content by UID without modifying, deleting, or executing external operations. However, severity is elevated to 'medium' rather than 'low' because email often contains sensitive personal information (credentials, financial details, private communications), so unrestricted reading could expose confidential data if an AI agent is compromised or misaligned.
From the tool's definition Tool description states 'Fetch the full content of an email' — a retrieval operation with no modification or deletion. The name 'read_email' and verb 'Fetch' align with Read category semantics.
Attacks that exploit this kind of access
Fetch the full content of an email by its UID (from search_emails results). It is categorised as a Read tool in the Claude Gmail MCP Server, which means it retrieves data without modifying state.
Register the Claude Gmail MCP server in PolicyLayer and add a rule for read_email: 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 Claude Gmail. Nothing to install.
read_email 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 read_email 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 read_email. 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.
read_email is provided by the Claude Gmail MCP server (pliablepixels/claude-gmail-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 →