AI agents use markAsUnread to create or update resources in Mail MCP Tool — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mail MCP Tool environment.
This tool modifies email state (unread flag) reversibly. It does not delete data (Destructive), execute code (Execute), send emails (Financial/Write severity depends on context), or move money. While it changes data, the change is reversible and low-impact, making it Write category with medium severity due to potential for metadata manipulation across multiple emails if chained with list operations.
From the tool's definition Tool name is 'markAsUnread' and the server description confirms this tool performs email operations. It belongs to a group of tools including 'markAsRead' and 'markMultipleAsUnread', which modify email metadata without deleting or sending messages.
Documented attack patterns abuse exactly the kind of access markAsUnread gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mail MCP Tool, and nothing reaches the server without passing your rules. This is the rule we recommend for markAsUnread:
{
"version": "1",
"default": "deny",
"tools": {
"markAsUnread": {
"limits": [
{
"counter": "markasunread_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}markAsUnread stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
markAsUnread. It is categorised as a Write tool in the Mail MCP Tool MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mail MCP Tool MCP server in PolicyLayer and add a rule for markAsUnread: 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 Mail MCP Tool. Nothing to install.
markAsUnread 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 markAsUnread 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 markAsUnread. 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.
markAsUnread is provided by the Mail MCP Tool MCP server (shuakami/mcp-mail). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 17 Mail MCP Tool tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
17 Mail MCP Tool tools catalogued and risk-classified — across an index of 42,500+ MCP servers.