Free disposable email inboxes for AI agents. Get verification codes and links. No API key needed.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Create a new disposable email inbox. Returns its address, password and a token for the other tools. Inboxes are receive-only and are kept until deleted. Save the address and password: tokens expire, and login_inbox gets a new one.
Optional domain. Defaults to the first available domain.
Optional part before the @. Random if omitted.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http mail https://freedomagentmail.us/mcp
Create a new disposable email inbox. Returns its address, password and a token for the other tools. Inboxes are receive-only and are kept until deleted. Save the address and password: tokens expire, and login_inbox gets a new one.
| Argument | Type | Description |
|---|---|---|
| domain | string | Optional domain. Defaults to the first available domain. |
| username | string | Optional part before the @. Random if omitted. |
{
"type": "object",
"properties": {
"domain": {
"type": "string",
"description": "Optional domain. Defaults to the first available domain."
},
"username": {
"type": "string",
"description": "Optional part before the @. Random if omitted."
}
}
}Delete an inbox and all of its email immediately.
| Argument | Type | Description |
|---|---|---|
| token* | string | The inbox token returned by create_inbox. |
{
"type": "object",
"required": [
"token"
],
"properties": {
"token": {
"type": "string",
"description": "The inbox token returned by create_inbox."
}
}
}List the most recent emails in an inbox (newest first), with codes, without full bodies.
| Argument | Type | Description |
|---|---|---|
| token* | string | The inbox token returned by create_inbox. |
{
"type": "object",
"required": [
"token"
],
"properties": {
"token": {
"type": "string",
"description": "The inbox token returned by create_inbox."
}
}
}Get a fresh token for an existing inbox (tokens expire; inboxes don't).
| Argument | Type | Description |
|---|---|---|
| address* | string | |
| password* | string |
{
"type": "object",
"required": [
"address",
"password"
],
"properties": {
"address": {
"type": "string"
},
"password": {
"type": "string"
}
}
}Read one email in full: plain text, codes, links and attachment names.
| Argument | Type | Description |
|---|---|---|
| token* | string | The inbox token returned by create_inbox. |
| message_id* | string |
{
"type": "object",
"required": [
"token",
"message_id"
],
"properties": {
"token": {
"type": "string",
"description": "The inbox token returned by create_inbox."
},
"message_id": {
"type": "string"
}
}
}Wait for unread email in an inbox and return it, including extracted one-time codes and verification links. Returns immediately if unread mail exists. Each message is returned once; it is marked read afterwards.
| Argument | Type | Description |
|---|---|---|
| token* | string | The inbox token returned by create_inbox. |
| timeout_seconds | number | How long to wait (0-60). Default 30. |
{
"type": "object",
"required": [
"token"
],
"properties": {
"token": {
"type": "string",
"description": "The inbox token returned by create_inbox."
},
"timeout_seconds": {
"type": "number",
"description": "How long to wait (0-60). Default 30."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 21, 2026.
Takes a password, API key or token as an argument, so the secret passes through the model's context and chat logs.
Fix: Read credentials from the server's own configuration (environment, OAuth) instead of a tool argument.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.