Mark one or more Gmail messages as unread
AI agents use mark_gmail_unread to create or update resources in Google Connections — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Google Connections environment.
This tool modifies email message state by changing the read/unread flag. It is a Write operation because it creates or modifies data reversibly. It is not Read (retrieves without modification), Execute (runs arbitrary code), Destructive (irreversible deletion), or Financial.
From the tool's definition Tool name 'mark_gmail_unread' and description 'Mark one or more Gmail messages as unread' indicate a state-modifying operation on Gmail messages.
Attacks that exploit this kind of access
Mark one or more Gmail messages as unread. It is categorised as a Write tool in the Google Connections MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Google Connections MCP server in PolicyLayer and add a rule for mark_gmail_unread: 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 Google Connections. Nothing to install.
mark_gmail_unread is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the mark_gmail_unread 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 mark_gmail_unread. 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.
mark_gmail_unread is provided by the Google Connections MCP server (michaelzrork/google-connections-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 →