AI agents use mark_as_read to create or update resources in IMAP MCP Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your IMAP MCP Server environment.
This tool modifies email state (read/unread flag) but does not create, delete, or destroy data. The change is fully reversible, making it a Write operation rather than Destructive. Severity is low because marking emails as read has minimal blast radius—it only affects UI/notification state without exposing sensitive data or causing irreversible harm.
From the tool's definition Tool name 'mark_as_read' and description 'Mark email as read' indicate a state change operation on email metadata. This is a reversible modification (can be marked unread again via the sibling tool 'mark_as_unread').
Documented attack patterns abuse exactly the kind of access mark_as_read gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and IMAP MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for mark_as_read:
{
"version": "1",
"default": "deny",
"tools": {
"mark_as_read": {
"limits": [
{
"counter": "mark_as_read_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}mark_as_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 email as read. It is categorised as a Write tool in the IMAP MCP Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the IMAP MCP Server MCP server in PolicyLayer and add a rule for mark_as_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 IMAP MCP Server. Nothing to install.
mark_as_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_as_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_as_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_as_read is provided by the IMAP MCP Server MCP server (non-dirty/imap-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 16 IMAP MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
16 IMAP MCP Server tools catalogued and risk-classified — across an index of 42,500+ MCP servers.