# send_message

Send a new email from an inbox.

Agent View of the PolicyLayer registry record for `send_message`. HTML page: https://policylayer.com/tools/dev-agenticemail-mcp/send-message

## Facts

- Tool: `send_message`
- Server: AgenticEmail (`https://api.agenticemail.dev/mcp`) — https://policylayer.com/tools/dev-agenticemail-mcp.md
- Homepage: https://github.com/https://api.agenticemail.dev/mcp
- Risk category: Write (Medium risk)
- Registry record: grade B, identity unverified
- Server auth posture: gated
- Server rate-limited: no
- Parameters: 5 (3 required)
- Recommended policy verdict: Rate-limited

## Parameters

| Parameter | Type | Required | Description |
| --- | --- | --- | --- |
| `to` | array | yes |  |
| `html` | string | no |  |
| `text` | string | no |  |
| `subject` | string | yes |  |
| `inbox_id` | string | yes |  |

Parameters from the server's own tool schema.

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "send_message",
    "arguments": {
      "to": [],
      "subject": "<subject>",
      "inbox_id": "<inbox_id>"
    }
  }
}
```

## Why send_message is rated Medium

Sending an email creates and transmits new data, which is a Write operation. It is not Read (does not merely retrieve data), not Execute (does not run arbitrary code or commands), not Destructive (emails sent can theoretically be unsent or are not permanently deleted), and not Financial (no money moves).

From the tool's own definition: "Tool name is 'send_message' and description states it will 'Send a new email from an inbox.' This creates new email data that is sent to external recipients."

Risk signals: Accepts raw HTML/template content (html)

## Use case

AI agents use send_message to create or update resources in AgenticEmail, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your AgenticEmail environment.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches AgenticEmail:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "send_message": {
      "limits": [
        {
          "counter": "send_message_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on AgenticEmail (8)

- `get_message` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/get-message.md
- `get_thread` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/get-thread.md
- `list_inboxes` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/list-inboxes.md
- `list_messages` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/list-messages.md
- `list_threads` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/list-threads.md
- `read_attachment` — Read — https://policylayer.com/tools/dev-agenticemail-mcp/read-attachment.md
- `create_inbox` — Write — https://policylayer.com/tools/dev-agenticemail-mcp/create-inbox.md
- `reply_to_message` — Write — https://policylayer.com/tools/dev-agenticemail-mcp/reply-to-message.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=dev-agenticemail-mcp · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/dev-agenticemail-mcp