AI agents use gmail_message_mark_unread to create or update resources in Mcp Gmail — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mcp Gmail environment.
Marking a message as unread is a reversible modification of message metadata/flags. It changes the read status of existing data without creating new content (not Write in the create sense) or deleting anything. This is less severe than Destructive actions but more impactful than pure Read operations.
From the tool's definition Tool description explicitly states "Mark a message as unread" — this modifies message state/metadata. The partial description cuts off but the name and available context indicate a state modification operation.
Attacks that exploit this kind of access
Mark a message as unread (add the. It is categorised as a Write tool in the Mcp Gmail MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mcp Gmail MCP server in PolicyLayer and add a rule for gmail_message_mark_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 Mcp Gmail. Nothing to install.
gmail_message_mark_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 gmail_message_mark_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 gmail_message_mark_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.
gmail_message_mark_unread is provided by the Mcp Gmail MCP server (knowledgeislands/mcp-gmail). 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 →