AI scans recent emails from meeting attendees and compiles briefing documents with open action items and conversation context.
Going into meetings without context on recent communications leads to redundant discussions, missed follow-ups, and wasted time. Manually searching your inbox for relevant threads before each meeting is impractical when you have five or more meetings a day.
MultiMail's AI agent scans your email history for recent conversations with upcoming meeting attendees and compiles concise briefing documents. Read_only mode means the agent never sends anything — it only reads and organizes, eliminating any risk of accidental communication.
Your AI agent receives your calendar events and identifies upcoming meetings with their attendee lists.
For each attendee, the agent uses search_contacts and check_inbox to find recent email exchanges, open threads, and pending action items.
The agent organizes findings into a structured brief: recent conversation summaries, open action items, pending decisions, and relevant context per attendee.
The brief is delivered to you 15-30 minutes before each meeting, giving you time to review context without digging through your inbox.
import requests
API = "https://api.multimail.dev/v1"
HEADERS = {"Authorization": "Bearer mm_live_xxx"}
def get_attendee_context(attendee_email: str):
"cm"># Search for the contact
contact = requests.get(
f"{API}/contacts/search",
headers=HEADERS,
params={"query": attendee_email}
).json()
"cm"># Check inbox for recent emails from this person
inbox = requests.get(
f"{API}/mailboxes/my-mailbox-id/inbox",
headers=HEADERS,
params={"limit": 10}
).json()
relevant = [e for e in inbox["emails"]
if attendee_email in str(e.get("from", ""))]
return relevant
"cm"># Build context for each meeting attendee
attendees = ["[email protected]", "[email protected]"]
for attendee in attendees:
emails = get_attendee_context(attendee)
print(f"\n{attendee}: {len(emails)} recent emails")
for e in emails:
print(f" {e[&"cm">#039;subject']}")Find recent email conversations with a meeting attendee using contact search.
import requests
API = "https://api.multimail.dev/v1"
HEADERS = {"Authorization": "Bearer mm_live_xxx"}
"cm"># Get full thread for an important conversation
thread = requests.get(
f"{API}/threads/{thread_id}",
headers=HEADERS
).json()
print(f"Thread: {thread[&"cm">#039;subject']}")
print(f"Participants: {&"cm">#039;, '.join(thread['participants'])}")
print(f"Messages: {len(thread[&"cm">#039;emails'])}")
# Read the latest message for current status
latest = requests.get(
f"{API}/emails/{thread[&"cm">#039;emails'][-1]['id']}",
headers=HEADERS
).json()
print(f"\nLatest from {latest[&"cm">#039;from']}:")
print(latest["text_body"][:300])Read full email threads to understand the current state of discussions.
"cm">// Search for each attendee's recent communications
const attendees = ["[email protected]", "[email protected]"];
for (const attendee of attendees) {
"cm">// Find contact details
const contact = await mcp.search_contacts({
query: attendee,
limit: 1
});
"cm">// Check inbox for recent exchanges
const inbox = await mcp.check_inbox({
mailbox_id: "my-mailbox-id",
limit: 20
});
"cm">// Get full thread context for ongoing conversations
for (const msg of inbox.emails) {
const email = await mcp.read_email({ email_id: msg.id });
if (email.from.includes(attendee)) {
const thread = await mcp.get_thread({
thread_id: email.thread_id
});
console.log(`Thread with ${attendee}: ${thread.subject}`);
}
}
}Use MCP tools to compile a meeting brief from email history.
AI compiles relevant email history across all attendees in seconds, replacing 15-20 minutes of manual inbox searching before each meeting.
The brief surfaces pending action items and unanswered questions from previous conversations, so nothing falls through the cracks.
The agent only reads your email — it never sends, replies, or modifies anything. Meeting briefs are compiled without any communication risk.
Walking in with full context means less time catching up and more time making decisions. Meetings become shorter and more productive.
Email infrastructure built for AI agents. Verifiable identity, graduated oversight, and a 38-tool MCP server. Formally verified in Lean 4.