Write Better Emails in a Fraction of the Time

AI drafts emails matching your style and tone. You review, refine, and approve before anything sends — combining AI speed with human judgment.


Why this matters

Writing professional emails consumes significant time, especially for difficult conversations, non-native language composition, or nuanced negotiations. Many professionals spend an hour or more daily composing and editing emails, time that could be better spent on strategic work.


How MultiMail solves this

MultiMail's AI agent composes email drafts based on natural language instructions, maintaining your writing style and adjusting tone for each context. With gated_send oversight, you review and refine every draft before delivery, keeping the human judgment that matters while offloading the writing effort.

1

Give Instructions

Tell your AI agent what you want to say, the tone to use, and any key points to include. Natural language instructions work: 'Reply to Marcus saying we're interested but need to see pricing first. Keep it professional but warm.'

2

AI Drafts the Email

The agent composes a draft using send_email or reply_email, matching your writing style learned from previous emails. It handles structure, tone, and formatting.

3

Review and Refine

With gated_send, the draft enters your approval queue. Read it, make edits, adjust tone, or add details the AI missed.

4

Send with Confidence

Approve the final version and it sends immediately. Track sent emails and thread replies for ongoing conversation management.


Implementation

Draft a New Email
python
import requests

API = "https://api.multimail.dev/v1"
HEADERS = {"Authorization": "Bearer mm_live_xxx"}

response = requests.post(
    f"{API}/send",
    headers=HEADERS,
    json={
        "from": "[email protected]",
        "to": ["[email protected]"],
        "subject": "Re: Partnership opportunity discussion",
        "text_body": (
            "Hi Marcus,\n\n"
            "Thank you for reaching out about the partnership opportunity. "
            "We&"cm">#039;d be interested in exploring how our platforms could work "
            "together.\n\n"
            "Before we dive deeper, could you share your pricing model "
            "and any technical requirements for the integration? That "
            "would help us evaluate the fit and timeline on our side.\n\n"
            "Looking forward to your thoughts.\n\n"
            "Best,\nYour Name"
        ),
        "html_body": "<p>Hi Marcus,</p><p>Thank you for reaching out...</p>"
    }
)
print(f"Draft queued for review: {response.json()[&"cm">#039;id']}")

Compose a professional email draft queued for review before sending.

Draft a Reply in an Existing Thread
python
import requests

API = "https://api.multimail.dev/v1"
HEADERS = {"Authorization": "Bearer mm_live_xxx"}

"cm"># Read the email we're replying to for context
original = requests.get(
    f"{API}/emails/{email_id}",
    headers=HEADERS
).json()

"cm"># Draft a reply in the thread
response = requests.post(
    f"{API}/reply",
    headers=HEADERS,
    json={
        "in_reply_to": original["id"],
        "text_body": (
            f"Hi {original[&"cm">#039;from'].split('<')[0].strip()},\n\n"
            "Thanks for sending over the pricing details. The Pro tier "
            "looks like a good fit for our needs.\n\n"
            "A few follow-up questions:\n"
            "1. Is there flexibility on the annual commitment?\n"
            "2. Can we start with a pilot period?\n"
            "3. What&"cm">#039;s the timeline for API access?\n\n"
            "Happy to jump on a call this week to discuss.\n\n"
            "Best regards"
        )
    }
)
print(f"Reply draft queued: {response.json()[&"cm">#039;id']}")

Reply to an email thread with an AI-drafted response.

MCP Tool: Email Drafting Workflow
typescript
"cm">// Read the email to reply to
const email = await mcp.read_email({
  email_id: originalId
});

"cm">// Get full thread for context
const thread = await mcp.get_thread({
  thread_id: email.thread_id
});

"cm">// Draft a reply (goes to approval queue under gated_send)
await mcp.reply_email({
  email_id: email.id,
  text_body: "Hi Marcus, thanks for the details. A few questions..."
});

"cm">// Check pending drafts awaiting approval
const pending = await mcp.list_pending({
  mailbox_id: "my-mailbox-id"
});
console.log(`${pending.emails.length} drafts awaiting your review`);

Use MCP tools to read context, draft replies, and manage the approval queue.


What you get

10x Faster Email Composition

AI drafts in seconds what takes you 10-15 minutes to write. Your time shifts from writing to reviewing — a much faster process.

Consistent Professional Tone

Every email maintains your professional voice regardless of time pressure, mood, or fatigue. The AI applies your style consistently.

Human Judgment Preserved

Gated send means you see and approve every email. AI handles the mechanics while you handle the nuance, strategy, and relationship awareness.

Better Non-Native Communication

For non-native English speakers, AI drafting produces polished, natural-sounding emails. Review mode lets you verify meaning while the AI handles grammar and idiom.


Recommended oversight mode

Recommended
gated_send
Outbound emails represent you personally and professionally. Gated send combines AI drafting speed with human review, ensuring every email reflects your judgment, voice, and relationship awareness before delivery.

Common questions

How does the AI learn my writing style?
You provide style examples and preferences to your AI agent — formal vs. casual, short vs. detailed, specific phrases you like. The agent also improves from your edits: when you modify a draft before approving, those changes inform future drafts.
Can I use different tones for different recipients?
Yes. Your AI agent adapts tone based on context and instructions. You might request a formal tone for client emails, casual for team communications, and diplomatic for negotiations. Recipient metadata in contacts can store tone preferences.
What about emails with sensitive content?
Gated send ensures you review every draft before delivery. For particularly sensitive emails — negotiations, HR matters, legal communications — you have full control to edit, rewrite, or reject the AI draft entirely.
Can the AI draft emails in multiple languages?
Yes. Modern AI models can compose fluent emails in dozens of languages. The AI drafts in your specified language while you review for accuracy. This is especially valuable for bilingual professionals managing communications across language boundaries.

Explore more use cases

The only agent email with a verifiable sender

Email infrastructure built for AI agents. Verifiable identity, graduated oversight, and a 38-tool MCP server. Formally verified in Lean 4.