Free
Try the API
- 3,000 events / month
- 1 agent
- TypeScript + Python SDK
- MCP server
- Magic-link signup, no card
No SSO/SAML, no procurement, no “contact us for Enterprise” — we don’t have a sales team to call. Free starts at 3K events/mo. Pro and Scale bundle send + inbound capacity. Metered (coming soon) bills $0.001/send + $0.002/inbound, +$0.003 opt-in classify.
Try the API
For solo agent builders
For shipping agent products
Coming soon — pay for what your agent uses
Sends are outbound emails. Inbound replies arrive as structured payloads with injection score + sender reputation always included. Classification (intent + entities extracted by our 2-call classifier) is opt-in — your agent probably runs an LLM anyway.
Pro covers your volume. Injection scan + reputation included on every inbound at the base price.
Get your API keyFor high-volume senders that want isolated reputation
Available on Scale and above (or on Metered with verified domain DNS). No KYC, no contract negotiation — available on request, provisioned per request.
The Metered tier bills three things separately. A send is one outbound email your agent dispatches through the API ($0.001 each). An inbound reply is one message delivered to one of your agents — includes parse to a structured payload, injection scan, sender-reputation lookup, and webhook delivery ($0.002 each). An optional classify step extracts intent + entities + urgency via our 2-call classifier (+$0.003 per inbound, opt-in per agent).
The split-rate matters: if your agent already runs an LLM on every reply, leave classification off and don't pay for work you're doing twice. If you want fast routing (switch (reply.intent)with no extra LLM call), turn it on and we'll do the extraction once for you.
On the Free tier, you hit a hard cap at 3,000 events per month (send + inbound combined). Paid tiers send a usage warning at 80% and let you finish the billing cycle without mid-month cutoffs.
All billing runs through Stripe. Monthly plans bill in advance with no annual commitments required. The Metered tier uses Stripe metered billing — each dimension (send / inbound / classify) is its own Subscription Item with its own usage record.
Your agent's email rides a managed sending layer with a documented multi-vendor failover runbook and a ~4‑hour DNS swap path, so a single provider outage doesn't take your agent offline. The full provider architecture is documented on the architecture page.
Every outbound email runs through a behavioral classifier before it leaves our system — rules plus a lightweight LLM call. Senders matching cold-prospecting patterns get auto-throttled to 50/day on shared IPs until reputation builds. No bans, no manual review, no account suspensions for borderline behavior. The classifier is invisible to your code — it's deliverability infrastructure, not a product axis you have to choose between.
A per-sender cron monitors complaint and bounce rates every 15 minutes. Any sender crossing 0.3% complaint rate gets paused before the upstream provider flags it at 0.5%. Your agent's deliverability isn't affected by someone else's behavior on the network. For high-volume senders that want reputation completely isolated from the shared pool, the Dedicated IP add-on ($50/mo) is available on request (Scale+) — provisioned per request, with ~14-day warmup.
Honest math vs the standard alternatives. Bursty agent workloads benefit most from per-event metered; high-volume steady workloads benefit most from monthly tiers. B2B prospecting is allowed under our AUP — same rules Resend follows (disclosure + unsub + company-email). Dedicated IP add-on $50/mo for reputation isolation.
Pricing accurate as of 2026-05. Phase 1 ships Free / Pro / Scale monthly tiers; Metered at $0.001/send + $0.002/inbound (+$0.003 opt-in classify) is coming soon. Dedicated IP add-on $50/mo on Scale+, available on request. See the full breakdown on /pricing or read the per-event metered rationale.
AgentMail provides programmatic agent inboxes — same primitive. Their gaps (per their own positioning): no native prompt-injection scanning, no per-agent reputation graph, no per-agent allowlists or budgets, monthly tiers with a $20→$200 jump and an Enterprise contact-sales gate at 150K/mo. We ship all of that, self-serve to any volume, with no sales team to call.
Architectural protections, not identity gates. (1) Free tier hard-capped at 3,000 events/month with email verification required. (2) Per-tier hourly + daily send caps in the billing layer. (3) Behavioral classifier auto-throttles cold-pattern senders. (4) Per-sender complaint cron auto-suspends anyone crossing a 0.3% rate — well before AWS's 0.5% threshold. (5) Per-agent reputation graph designed to propagate suppression network-wide as cohort 1 grows — a banned sender loses verified-agent status across every account that has joined. No KYC, no government ID upload — the same friction floor as Resend and the rest of the transactional category.
Multi-vendor failover is documented as a launch-hardening requirement — a runbook to stand up backup senders (Postmark, SparkPost, Brevo) with a ~4-hour DNS swap path. The auto-suspend system already monitors per-sender complaint and bounce continuously, pausing anyone crossing 0.3% before AWS sees 0.5%, so reputation damage is contained well before a vendor escalates.
Public API opens Q3 2026. Drop ~6 lines into your agent and ship.
$ npm install @mailsai/sdk