Delete a Gmail draft message
AI agents call deleteDraft to permanently remove resources in Gmail MCP Server — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Deleting a draft is an irreversible action that removes user data without possibility of recovery through the tool itself. While the blast radius is somewhat contained (only affects drafts, not sent mail or other critical data), the destructive nature and potential for accidental loss of user work justifies 'high' severity. Confidence is high because the action is unambiguously destructive.
From the tool's definition The tool name is 'deleteDraft' and description states 'Delete a Gmail draft message' — the verb 'delete' indicates irreversible removal of data.
Attacks that exploit this kind of access
Delete a Gmail draft message. It is categorised as a Destructive tool in the Gmail MCP Server MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Gmail MCP Server MCP server in PolicyLayer and add a rule for deleteDraft: 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 MCP Server. Nothing to install.
deleteDraft is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the deleteDraft 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 deleteDraft. 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.
deleteDraft is provided by the Gmail MCP Server MCP server (zacco16/gmail-mcp-server). 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 →