AI agents use compose_email to create or update resources in Eml — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Eml environment.
compose_email creates a new email draft file (.eml), which is a reversible write operation. While email composition could potentially be misused (e.g., crafting phishing or spam), the action itself is write-class (creates data). Severity is medium because misuse could enable fraudulent or harassing communications, but the impact is limited to draft creation without automatic sending or financial consequences.
From the tool's definition Tool description explicitly states 'Create a new .eml draft', which is a write operation that creates new data. The sibling tools include delete_email (destructive) and update_email (write), contextualizing this as a data creation action.
Attacks that exploit this kind of access
Create a new .eml draft and open it in the default mail client. It is categorised as a Write tool in the Eml MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Eml MCP server in PolicyLayer and add a rule for compose_email: 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 Eml. Nothing to install.
compose_email 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 compose_email 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 compose_email. 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.
compose_email is provided by the Eml MCP server (miguelripoll23/eml-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 →