send_email

THE RISK

High Risk

Why send_email needs a policy

Sending an email is an external operation that cannot be trivially undone once delivered. It interacts with external systems and recipients. While not financial or destructive, it executes an external action (delivering a message to a customer) whose effects depend on the content and recipient arguments. The 'Requires governance approval' note confirms its recognized risk level.

From the tool's definition 'Send an email to a customer' — triggers an external communication operation with real-world side effects

Documented attack patterns abuse exactly the kind of access send_email gives an agent:

POLICY

How to control send_email

PolicyLayer is an MCP gateway — it sits between your AI agents and Sdk, and nothing reaches the server without passing your rules. This is the rule we recommend for send_email:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "send_email": {
      "limits": [
        {
          "counter": "send_email_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

send_email stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Sdk — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Sdk tools

Execute governance_proxy SidClaw MCP Governance Proxy — wraps any upstream MCP server with policy evaluation, Write sidclaw_record Log the outcome of an action that has already executed. Use this when you don Write sidclaw_session_end Close out a client-side session created via sidclaw_session_start. Not Other sidclaw_session_start Generate a client-side session ID for your own bookkeeping. SidClaw does Read search_docs Search the internal knowledge base for documentation and policies. Read sidclaw_policies List active policy rules for the current tenant (or a specific agent). Read export_data Export customer data records. Blocked by data protection policy. Other sidclaw_approve Wait for a human approval to resolve. Use after sidclaw_evaluate returns

All 10 Sdk tools →

Execute tools on other servers

Android Forensics ADB MCP Server adb_shell_command EMBA-MCP run_emba_scan Frida Game Hacking MCP remove_breakpoint Procmon request_elevation

Go deeper

High-risk tools in Sdk → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Agent sandbox → Bounding what an agent can execute and where the boundary should sit.
Least privilege for MCP → Granting agents the minimum tool surface the task needs.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about send_email

What does the send_email tool do? +

Send an email to a customer. Requires governance approval. It is categorised as a Execute tool in the Sdk MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on send_email? +

Register the Sdk MCP server in PolicyLayer and add a rule for send_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 Sdk. Nothing to install.

What risk level is send_email? +

send_email is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit send_email? +

Yes. Add a rate_limit block to the send_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.

How do I block send_email completely? +

Set action: deny in the PolicyLayer policy for send_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.

What MCP server provides send_email? +

send_email is provided by the Sdk MCP server (@sidclaw/sdk). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Sdk tool call.

Start from Sdk, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →