"How to allow an LLM to ask questions to a user" matching MCP servers:

One-pass agentic inbox triage as an MCP server: fetch unread Gmail → classify (action_needed/fyi/newsletter/noise) → summarize → extract tasks → draft replies as Gmail DRAFTS (never sends) → flag calendar → write a triage report. Four stdio tools (fetch_emails, save_gmail_draft, append_tasks, write_report); the host is the LLM, so it runs keyless in Claude Code. Gmail scopes: readonly + compose

Sends document files (EPUB, PDF, DOCX, etc.) to your Kindle via email using Amazon's Send to Kindle feature.

A Model Context Protocol (MCP) server that allows Claude Desktop to send documents directly to your Kindle e-reader.

An MCP server that sends plain-text Gmail messages only to a predefined allowlist, using generated recipient IDs instead of email addresses to enforce security.

Enables pushing enriched leads into Instantly or Smartlead campaigns with ICP score gating, deduplication, and dry-run previews via an Apify actor.

Official MCP server for Shipmail, enabling agents to manage domains, mailboxes, messages, threads, webhooks, and suppressions via natural language.

Enables sending documents (Markdown, PDF, EPUB) directly to a Kindle via Amazon Send-to-Kindle email, with Markdown converted to EPUB using pandoc.

MCP server for Yahoo Mail over IMAP+SMTP that enables AI assistants to manage emails fully—listing, reading, searching, threading, flagging, moving, archiving, handling attachments, and sending/reply/forward/draft—using an app-specific password.

Multi-user Gmail MCP server that enables sending/receiving emails, managing accounts, and searching mail via SMTP/IMAP with per-user isolation and encrypted app passwords.

Enables AI assistants to interact with Microsoft 365 and Office services through the Microsoft Graph API, providing tools for email, calendar, files, Teams, SharePoint, and more with configurable permissions and output formats.

Enables AI agents to send emails, create drafts in Gmail, and append content to Google Docs via standardized MCP tools.

Enables AI agents to read and send email, handle attachments and manage calendar invitations through their own IMAP/SMTP mailboxes, with per-mailbox sender and recipient allow lists and automatic HTML-to-Markdown conversion. It runs as a single self-hosted Docker container, exposing the same tools over MCP and REST.

An MCP server that exposes a local notmuch email database to an LLM client such as Claude. It is read-first: searching, reading, and understanding mail is always available; writing anything (drafts, tags, exported files) requires an explicit opt-in flag and is confined to clearly bounded locations.

▎ Check an email before an AI agent sends it: send / ask a human / block — with machine reason codes and prompt-injection detection.

Read and write email through the Atomic Mail from an AI agent. Handles proof-of-work authentication and JMAP so the agent thinks in JMAP method calls. Use when the user asks to register an email inbox, list mailboxes, fetch or send email.

Enables an LLM client to read a structured CV resource and send tailored application emails to recruiters via SMTP.

Lets Claude connect to a Yahoo Mail account with an app password to read messages, create and manage folders, move messages, and compose or send email, with an optional read-only mode.

MCP server for SentVia that provides email infrastructure for AI agents, enabling them to create inboxes, send, reply, forward, search messages, manage drafts, domains, webhooks, and allow/block rules through 21 tools.

Enables MCP-compatible AI clients to securely read a user's Microsoft 365 data across Outlook, Calendar, Teams, meetings, OneNote, SharePoint, OneDrive, Loop, and Search, and to send Outlook email only after explicit user confirmation.

Connects AI assistants to Masaryk University's information system in read-only mode, letting students ask about their timetable, deadlines, emails, grades, and study materials in natural language.