Back to MCP Directory
Inbox Zero
Model Context Protocol moderate risk

Inbox Zero

Inbox Zero is an open-source AI email assistant that organizes your inbox, drafts replies, and tracks follow-ups, aiming to help users achieve inbox zero faster.

Connections & Capabilities

Connects To

GitHubPostgreSQLRedisDiscordDockerGoogle DriveVercel

Capabilities

readwriteexecadmin

Quickstart

Install

npx @inbox-zero/cli setup      # One-time setup wizard

Exposed MCP Tools (4)

moderate
bulk_unsubscriber

Unsubscribes and archives emails in bulk.

Modifies email subscriptions, but does not delete data.

moderate
bulk_archiver

Archives old emails in bulk to clean up the inbox.

Archives emails, potentially making them less accessible.

moderate
ai_email_drafting

Pre-drafts email replies using AI.

Generates email content, which could contain errors or unintended disclosures.

moderate
smart_filing

Automatically saves email attachments to cloud storage.

Moves files to cloud storage, which could have privacy implications.

Safety Assessment

Inbox Zero offers helpful tools for email management, but its access to email accounts and reliance on AI introduces moderate risks. Self-hosting provides more control but requires careful security practices. Using the hosted version relies on the security of the hosting provider.

  • Open-source, allowing for community review and scrutiny.
  • Offers self-hosting options, providing greater control over data and infrastructure.
  • Uses OAuth for authentication, reducing the risk of exposed credentials.
  • Provides bulk unsubscribe and archiver tools to manage email effectively.
  • Requires access to email accounts, potentially exposing sensitive information.
  • Relies on AI to draft replies, which could lead to unintended disclosures or misrepresentations.
  • Self-hosting requires careful configuration and maintenance to avoid security vulnerabilities.
  • Configuration requires secrets management for API keys and database credentials.
Inbox Zero Review, Pricing & Alternatives (2026) | Flaex AI