Skip to content
agentboxd
  • Features
  • How it works
  • Docs
  • Guides
  • MCP
  • Pricing
Sign inGet an API key
  • Features
  • How it works
  • Docs
  • Guides
  • MCP
  • Pricing
  • Sign in

Latest 25 Sept 2026 · RSS

Changelog.

What we shipped, newest first. Everything here is live on agentboxd.com; each entry links to its docs. The full list of what Agentboxd does is on Features.

2026-09-25

Agentboxd, drafts, realtime and agent sign-in

25 Sept 2026 · 12 changes

  • Agent self-signup

    An agent with no API key can create its own workspace and inbox: Agentboxd.signup() in both SDKs, or the signup tool when the MCP server starts without a key. Until a person claims it from the emailed link, it replies freely but emails at most 20 new recipients a day.

    Read the docs: Agent self-signup
  • Sign in with Agentboxd

    Every inbox is now also an identity. Your agent asks for a 5-minute, single-use ID token and signs in to apps that accept it, over standard OpenID Connect at id.agentboxd.com. Apps register in the new Identity page of the dashboard.

    Read the docs: Sign in with Agentboxd
  • Drafts, approval and scheduled send

    Agents can write drafts that a person approves in the inbox’s Drafts tab, and any draft or send can be scheduled 1 minute to 30 days ahead with send_at. Six new MCP tools, and drafts in both SDKs.

    Read the docs: Drafts, approval and scheduled send
  • Realtime event stream

    The webhook events over a WebSocket at /v1/stream, with replay after a disconnect. No public URL needed. mr.stream() in the TypeScript SDK and AsyncAgentboxd.stream() in Python.

    Read the docs: Realtime event stream
  • SMTP submission

    Send from any SMTP client on smtp.agentboxd.com, ports 587 and 465, with an API key as the password. Messages go through the same checks and delivery as the API.

    Read the docs: SMTP submission
  • Plans, retention and a burst limit

    Free, Builder, Team and Scale plans measured in emails and AI calls, with overage on paid plans. Mail is kept for the plan’s retention (the keep label opts a message out), and a 5-minute send limit stops runaway loops.

    Read more: Plans, retention and a burst limit
  • Dashboard v2

    Scoped API keys with permission presets, allow and block lists, a metrics overview, a webhook event catalog with envelope (no-body) payloads and 24-hour delivery stats, and a “Use the API” button with the calls for the page you’re on.

    Read the docs: Dashboard v2
  • Security page, DPA and sub-processors

    What protects your agents’ mail today and what is still planned, a GDPR Data Processing Addendum, the list of sub-processors, and /.well-known/security.txt.

    Read more: Security page, DPA and sub-processors
  • Hosting and backups, in detail

    The service runs in France (EU). Nightly backups are now encrypted before they leave the server, kept on a second EU server, and restore-tested; health checks run every 5 minutes.

    Read more: Hosting and backups, in detail
  • Webhook SSRF guard

    Webhook URLs must be public HTTPS endpoints. Hostnames are checked when you save a webhook and again before every delivery, and redirects are not followed.

    Read the docs: Webhook SSRF guard
  • Guides, use cases and blog

    Step-by-step guides for MCP clients, LangChain, CrewAI, the OpenAI Agents SDK, the Vercel AI SDK, n8n and Playwright, each with a runnable example.

    Read more: Guides, use cases and blog
  • Now called Agentboxd

    New name, same service. The packages are agentboxd on npm and PyPI and @agentboxd/mcp, and the SDKs read AGENTBOXD_API_KEY (the old variable names still work).

    Read the docs: Now called Agentboxd

2026-09-24

Launch

24 Sept 2026 · 7 changes

  • Real inboxes for AI agents

    Create an inbox with one API call and receive mail on our own MX, with SPF, DKIM and DMARC checks, threading, attachments and extracted_text. Agents send and reply DKIM-signed from their own address.

    Read the docs: Real inboxes for AI agents
  • Verification codes

    waitForVerification returns the login code or magic link from a sign-up email with a confidence score.

    Read the docs: Verification codes
  • AI triage and the privacy switch

    JEV categorises every inbound message and scores prompt injection, phishing, needs-a-human, auto-reply and urgency. One workspace setting (off, categorize or full) decides what content may reach a model.

    Read the docs: AI triage and the privacy switch
  • Contacts, knowledge and reply drafts

    A contact per sender with notes and metadata, searchable knowledge documents, and DeepSeek reply drafts that cite them and are never sent on their own.

    Read the docs: Contacts, knowledge and reply drafts
  • Custom domains and temporary inboxes

    Send and receive on your own subdomain with its own DKIM key, and create receive-only inboxes on tmp.agentboxd.com that wipe themselves when they expire.

    Read the docs: Custom domains and temporary inboxes
  • MCP server and SDKs

    An MCP server for Claude Desktop, Claude Code and Cursor, and TypeScript and Python SDKs.

    Read the docs: MCP server and SDKs
  • Webhooks and search

    Signed webhooks with retries, long-poll wait endpoints and full-text search.

    Read the docs: Webhooks and search
agentboxd

Email inboxes for AI agents. Hosted in the EU, or in your own cloud on Enterprise.

MX
mx.agentboxd.com
API
api.agentboxd.com

Product

  • Features
  • How it works
  • Verification codes
  • Sign in with Agentboxd
  • Pricing
  • Changelog

Trust

  • Security
  • Privacy and data flow
  • Sub-processors
  • DPA
  • Report a vulnerability

Docs

  • Quickstart
  • API reference
  • Webhooks
  • Realtime events
  • SMTP
  • MCP server

Resources

  • Guides
  • Use cases
  • Blog
  • Claude and Cursor setup
  • Testing sign-up emails

SDKs

  • TypeScript
  • Python
  • Enterprise

Account

  • Sign in
  • Dashboard
  • Support
  • Contact
  • Report abuse
© 2026 AgentboxdPrivacySecurityData hosted in France (EU)Free plan · no card