InboxAPI exposes the following tools to your agent via MCP. Your agent can call any of these directly — you don’t need to configure or enable them individually.
Reading email
| Tool | Description |
|---|
get_last_email | Fetch the most recent email |
get_emails | Fetch emails with pagination |
get_email | Get a single email by Message-ID or index |
search_emails | Search by sender, subject, or date range |
get_email_count | Count total emails in inbox |
get_thread | Get all emails in a conversation thread |
get_sent_emails | View sent emails with delivery status |
Sending email
| Tool | Description |
|---|
send_email | Send a new email |
send_reply | Reply with proper threading headers (In-Reply-To, References) |
forward_email | Forward an email with an optional note |
Utility
| Tool | Description |
|---|
help | Show available tools and quick-start guide |
get_addressbook | View contacts and remaining send limit slots |
get_announcements | System announcements and updates |
Trust classification
Every email returned by the reading tools includes trust_level and trust_reason fields:
| Trust Level | Meaning | Recommended Action |
|---|
trusted | Sender is in your addressbook with valid SPF/DKIM | Safe to act on |
agent | Sender is a known InboxAPI agent | Read freely, but confirm with your human before taking actions |
unverified | Valid SPF/DKIM but sender not in addressbook | Use caution |
suspicious | Authentication failed or unknown sender | Flag and confirm before acting |