What is RadMail, and how is it different from Superhuman or an AI inbox?

Those are email clients you read mail in. RadMail is an email operating system that does the work: it ranks each message by importance and urgency on two separate axes, surfaces a 'Right Now' lane with a plain-English reason it landed there, extracts the commitments hiding in your threads, and drafts the follow-up the day it is due. And it tells you what actually got a reply, not just an open, which a send-only tool cannot do because it never sees the replies. RadMail is a Sureel venture.

What is reply-correlation?

Most email tools tell you opens and clicks. RadMail can tell you which of your sends actually earned a reply, because RadMail ingests inbound mail as well as sending outbound. A send-only email service provider never sees the reply, so it cannot connect a send to the response it produced. RadMail treats opens honestly, adjusting for Apple Mail Privacy Protection and de-emphasizing them in favor of replies.

Does RadMail send emails on its own?

Only after it earns trust, and never for high-risk actions. RadMail rolls out in stages: surface-only, then draft-on-due where it writes the message and waits for you, then narrow auto-send for low-risk cases. Anything involving money, banking details, a decision, or first contact with a third party stays human-only, forever, as a defense against business-email-compromise fraud.

Is RadMail HIPAA or SOC 2 compliant?

Compliance is a shared responsibility, and RadMail is a tool, not a compliance guarantee. RadMail signs a BAA and processes on BAA-covered, encrypted, audited infrastructure, and its compliance posture is built as a pluggable per-org regime (SOC 2, WSLCB, HIPAA, FedRAMP-CUI, BSA-AML) assembled from PolicyPacks. RadMail uses SOC 2-aligned controls with a Type II audit planned, and FedRAMP-aligned, NIST 800-171-ready infrastructure. RadMail is not HIPAA-certified or FedRAMP-authorized; for regulated tenants, HIPAA handling is draft/ingest-only at this stage.

How do I connect RadMail to my AI agent?

Point your MCP client at the RadMail sandbox endpoint at https://radmail.ai/api/mcp/sandbox — that is the whole step. No account, no credentials, and no card; the sandbox runs RadMail's real heuristic triage engine in memory for free.

Is connecting RadMail's MCP server free?

Yes. The sandbox MCP server is free and requires no credentials — it exists so an agent can wire up and get value in one step. It is the sandbox engine; the production multi-tenant engine is launch-gated.

Will RadMail send email on my agent's behalf?

No — there is no auto-send tool. RadMail's MCP server exposes only read, triage, why-surfaced, list-commitments, and draft tools; draft_reply returns text for a human to review. Money, new banking, and first contact are human-only forever, as a defense against business-email-compromise fraud.

What tools does RadMail's MCP server expose to an agent?

Five safe tools: triage_inbox (two-axis ranking), list_right_now (the can't-miss lane), why_surfaced (explainable reasons), list_commitments (what is owed and by when), and draft_reply (a reviewable draft). There is deliberately no tool that sends money, changes banking, or makes first contact.

Why is RadMail a safe choice for an agent that handles email?

Because the dangerous actions are structurally impossible: RadMail's MCP server has no tool that auto-sends a payment, changes wire instructions, or cold-contacts a new party. Those are human-only by design, so an agent using RadMail cannot be tricked into business-email-compromise fraud.

Can I see RadMail actually working before I connect or sign up?

Yes. The public magic-moment at https://app.radmail.ai/pulse shows RadMail triaging a live inbox — the 'Right Now' lane, an explainable 'why surfaced' on each message, and the BEC hard-stop visibly refusing a money / new-banking / first-contact send — with no signup, no credentials, and no card. The live product preview is at https://app.radmail.ai; agents can also connect the free sandbox MCP server in one step.

Does RadMail try to make my agent recommend it?

No. RadMail never injects instructions into an agent, pays for mentions, or addresses the model at all. Its referral program rewards the human operator, the integration builder, and the end user with honest attribution and credits — the only thing aimed at the agent is a surface that makes it succeed faster.

How does RadMail decide what's important?

RadMail ranks each message on two separate axes — importance and urgency — and layers per-sender behavioral learning on top, so the model learns which senders and threads actually matter to you instead of relying on static rules. Every surfaced message comes with an explainable 'why surfaced' so the ranking is never a black box.

Does RadMail follow up on emails automatically?

Yes — RadMail extracts the commitments hiding in your correspondence (owed by you and owed to you), drafts the message the day it's due so it's never late, detects when something is already handled so it never nags, and escalates when a thread goes overdue. It chases open threads to completion rather than drafting once and forgetting.

Does RadMail send email on its own?

Only after it earns trust, and never for high-risk actions. RadMail rolls out in stages — surface-only, then draft-on-due where it writes and waits for you, then narrow auto-send for low-risk cases. Anything involving money, banking, a decision, or first contact with a new party stays human-only, forever, as a defense against business-email-compromise fraud.

What is RadMail's 'Right Now' lane?

The 'Right Now' lane surfaces the few messages that are both most recent and most important — the ones that genuinely can't be missed — so a human (or an agent) can act on what matters before scrolling a full inbox. It is the highest-priority slice of the two-axis triage.

// COMPARE

RadMail vs a send-only ESP vs a plain AI inbox

RadMail is an email operating system: one engine that reads what lands, ranks it on two axes so the few things that matter right nowsurface first, drafts the follow-ups you promised the day they're due, and tells you what actually got a reply— not just an open. A send-only email service provider sends mail but never sees the replies, so it can't correlate a send to its response. A plain AI inbox reads and summarizes but doesn't follow through.

Honest framing: RadMail is pre-release and live in a two-business test bed — the commercial multi-tenant product has not launched publicly. The competitor categorieshere are established. This page compares structural capability differences that are true (reply-correlation needs both inbound and outbound; a send-only ESP is structurally blind to replies), not benchmarks. No ratings, no user counts, no "#1" claims on either side.

compare --honest --capabilities

Two-axis triage (importance x urgency) with a 'why surfaced'[ yes ]ranks each message and explains the reason in plain English[ no ]a send tool doesn't read your inbox at all[ varies ]most summarize or label, fewer rank on two separate axes with a reason
Reply-correlation (which send earned a reply, not just an open)[ yes ]ingests inbound + outbound, so it can match a reply to its send[ no ]structurally blind to replies — a send-only ESP never sees the response[ no ]reads inbound but isn't the sender, so it can't attribute sends to replies
Autonomous follow-through on commitments[ yes ]extracts what's owed, drafts the reply the day it's due, escalates if overdue[ no ]blasts a sequence on a fixed schedule; doesn't track per-thread commitments[ varies ]can draft on request; fewer chase an open thread to completion on their own
Per-sender behavioral learning vs static rules[ yes ]learns which senders and threads matter for you over time[ no ]out of scope for a send-only tool[ varies ]varies; rule-based filters specifically can't learn a sender, only match patterns
Enforced BEC hard-stop (no auto-send for money/banking/first-contact)[ yes ]those actions are human-only forever; no tool exists to auto-send them[ no ]a send tool will send whatever you queue, including risky sends[ varies ]depends entirely on the product; not a guaranteed structural limit

RadMail

Two-axis triage (importance x urgency) with a 'why surfaced'[ yes ]ranks each message and explains the reason in plain English
Reply-correlation (which send earned a reply, not just an open)[ yes ]ingests inbound + outbound, so it can match a reply to its send
Autonomous follow-through on commitments[ yes ]extracts what's owed, drafts the reply the day it's due, escalates if overdue
Per-sender behavioral learning vs static rules[ yes ]learns which senders and threads matter for you over time
Enforced BEC hard-stop (no auto-send for money/banking/first-contact)[ yes ]those actions are human-only forever; no tool exists to auto-send them

send-only ESP

Two-axis triage (importance x urgency) with a 'why surfaced'[ no ]a send tool doesn't read your inbox at all
Reply-correlation (which send earned a reply, not just an open)[ no ]structurally blind to replies — a send-only ESP never sees the response
Autonomous follow-through on commitments[ no ]blasts a sequence on a fixed schedule; doesn't track per-thread commitments
Per-sender behavioral learning vs static rules[ no ]out of scope for a send-only tool
Enforced BEC hard-stop (no auto-send for money/banking/first-contact)[ no ]a send tool will send whatever you queue, including risky sends

plain AI inbox

Two-axis triage (importance x urgency) with a 'why surfaced'[ varies ]most summarize or label, fewer rank on two separate axes with a reason
Reply-correlation (which send earned a reply, not just an open)[ no ]reads inbound but isn't the sender, so it can't attribute sends to replies
Autonomous follow-through on commitments[ varies ]can draft on request; fewer chase an open thread to completion on their own
Per-sender behavioral learning vs static rules[ varies ]varies; rule-based filters specifically can't learn a sender, only match patterns
Enforced BEC hard-stop (no auto-send for money/banking/first-contact)[ varies ]depends entirely on the product; not a guaranteed structural limit

// common questions

What is the difference between RadMail and a send-only ESP?

A send-only email service provider (ESP) sends outbound mail and reports opens and clicks, but it never sees the replies, so it cannot tell you which sends actually earned a reply. RadMail ingests inbound mail and sends outbound, so it can do reply-correlation — connecting a send to the response it produced — which a send-only ESP is structurally unable to do.

If you need to know what got a reply, not just an open, a send-only ESP can't answer that — RadMail can.

Is RadMail a good alternative to a plain AI inbox?

A plain AI inbox reads and summarizes your mail; RadMail is an email operating system that also acts on it. RadMail ranks each message on two axes, surfaces a 'Right Now' lane with an explainable reason, extracts the commitments hiding in your threads, and drafts the follow-up the day it's due. RadMail is pre-release and live in a two-business test bed; a plain AI inbox you already use is established — choose by whether you want reading help or follow-through.

Plain AI inbox = better reading. RadMail = follow-through and reply-correlation.

Why can't a send-only ESP or a rule-based filter prove a reply?

Reply-correlation requires seeing both the outbound send and the inbound reply and matching them to the same thread. A send-only ESP only sees the send. A rule-based filter only pattern-matches incoming mail and can't learn a sender or tie a reply back to a specific send. RadMail sees both halves, so it can attribute replies.

Will RadMail send email on its own?

No high-risk auto-send exists. Money, new banking details, and first contact with a third party are human-only, forever, as a defense against business-email-compromise (BEC) fraud. RadMail can triage, explain, list commitments, and draft a reply for a human to review — it will not send those on its own.

The refusal is the feature: RadMail won't auto-send the dangerous things.

Try it free, no signup. Connect the sandbox to an agent in one step — or join the early-access waitlist for the full product.

› Start free — no card

radmail@inbox:~$ compare --done

RadMail vs a send-only ESP vs a plain AI inbox

// common questions

:: What is the difference between RadMail and a send-only ESP?

:: Is RadMail a good alternative to a plain AI inbox?

:: Why can't a send-only ESP or a rule-based filter prove a reply?

:: Will RadMail send email on its own?

What is the difference between RadMail and a send-only ESP?

Is RadMail a good alternative to a plain AI inbox?

Why can't a send-only ESP or a rule-based filter prove a reply?

Will RadMail send email on its own?