AI agents use mark_email_read to create or update resources in Fastmail MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Fastmail MCP Server environment.
This tool modifies email state (read/unread flag) but is fully reversible: an email can be marked unread again. It affects only metadata, not content, and has no destructive or financial impact. The blast radius is minimal—a compromised agent could change read flags but cannot delete emails, access sensitive content irreversibly, or cause harm beyond minor organizational confusion. This is a clear Write operation.
From the tool's definition Tool name is 'mark_email_read' and description states 'Mark an email as read or unread' — a state change operation that modifies email metadata reversibly.
Documented attack patterns abuse exactly the kind of access mark_email_read gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Fastmail MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for mark_email_read:
{
"version": "1",
"default": "deny",
"tools": {
"mark_email_read": {
"limits": [
{
"counter": "mark_email_read_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}mark_email_read 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.
Mark an email as read or unread. It is categorised as a Write tool in the Fastmail MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Fastmail MCP Server MCP server in PolicyLayer and add a rule for mark_email_read: 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 Fastmail MCP Server. Nothing to install.
mark_email_read 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_email_read 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_email_read. 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_email_read is provided by the Fastmail MCP Server MCP server (madllama25/fastmail-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Fastmail MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
38 Fastmail MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.