Skip to content

Real-time email verification

Stop paying for emailsthat don't exist.

PricklyMails checks every address before you send. Keep the good ones, drop the rest.

Try it on your address

Can your address receive email?

Free, no card required. An answer in a few seconds.

More than 50 tools plug straight in
  • Gmail
  • Brevo
  • Mailchimp
  • Mailjet
  • Outlook
  • Resend
  • lemlist
  • Klaviyo
  • Mailgun
  • Kit
  • Loops
  • Campaign Monitor
  • Substack
  • SparkPost
  • Mailtrap
  • Proton Mail
  • Ghost
  • HEY
  • iCloud
  • Thunderbird
The problem

Too many bounces, and it's no longer the address that gets blocked. It's you.

Amazon SES puts an account under review at a 5% bounce rate and can pause its sending at 10%. Long before that, a dead address is already costing you, somewhere that depends on what you send. Pick yours.

You send to a list

A few dead addresses are enough to get your sending suspended.

Sending platforms watch your bounces: Brevo suspends an account above 2% on a send to new contacts, HubSpot at 5% in a month. Each bounce also wears down your sender reputation, the one that decides whether the rest of your mail arrives.

400bounces from a single send to 10,000 new contacts with 4% dead addresses. Brevo suspends an account past 200.

Your list

a.kowalski@example.com

contact@example.org

priya.shah@example.com

j.okafor@example.net

info@old-site.example

s.tanaka@example.com

lucia.romero@example.org

m.dubois@example.net

Each one counts against your account.
What we check

We check what can be checked, and tell you about the rest.

  • The receiving server, asked live.

    We ask it whether the mailbox exists, without sending an email. Google, Microsoft and Yahoo don't answer that question reliably: we say so, rather than show a confirmation they never gave.

  • Dead, parked and throwaway domains, set aside.

    A domain with no mail server, one that refuses all email or sits parked, a throwaway address, even one hiding behind a disposable service's mail servers: we tell you to remove it before you send.

  • Typos caught, not rejected.

    gmial.com? You get gmail.com as the suggested fix, before the email ever bounces.

  • A clear call for every address.

    Send, review or remove, with the reason spelled out and a 0 to 100 score for your own rules. A shared inbox like contact@ or a domain that accepts everything goes to review.

See it in action

You know who to keep, before you hit send.

Every address comes back with a clear call: send, review, or remove. Here is a sample list of 1,520 contacts, as your dashboard shows it.

Pricing

You only pay for what we actually verify.

Pick a monthly plan and choose which checks to run. You pay for the checks, not the email count.

Starter
€8/mo
~1,300 verifications
To get started, with webhooks and higher limitsEverything in Free, plus:
  • 1 AI agent over MCP
  • 1 webhook
  • Lists of up to 1,000 addresses

Compare all plans

FreeStarterProBusinessEnterprise
Price€0€8 /mo€32 /mo€120 /mo€400 /mo
Tokens / mo1,000 (one-time)10,00050,000250,0001,000,000
Verifications~130~1,300~6,300~31,000~130,000
Requests / min12603001,0005,000
Max batch1001,00010,000100,0001,000,000
API keys131050100
Agents1 (verify + read only)14812
Available checks66777
Webhooks-12410
SupportStandardStandardStandardPriorityPriority

Need more? Top up at €0.001 per token. Purchased tokens never expire.

By the numbers

Fast where it can be, patient where it has to be.

Emails verified
For 9 verdicts out of 10

Google, Microsoft and Yahoo addresses are settled without asking their server, so the verdict is instant. For the rest, we ask the receiving server, which takes a few seconds. Real figures, updated daily.

How to use it

One engine, four ways in.

The same engine behind every way in, paid for with the same tokens.

  • Dashboard and batches

    Whole lists at once.

    Drop a CSV into the dashboard, or send the list through the API: from 100 to 1,000,000 addresses per batch depending on your plan, deduplicated, and a report with the call for every address.

  • REST API

    The verdict, as JSON.

    Submit an address at sign-up or from your CRM, read the verdict as soon as it's ready. A free test key returns fixed results, so you can wire everything up without spending a token.

  • Webhooks

    Results come to you.

    A signed webhook (HMAC-SHA256) for every finished verification. Test your receiver in one click and follow every delivery.

  • AI agents

    Your agent checks on its own.

    Our MCP server gives Claude, or any compatible client, the tools to verify one address or a whole list. It signs in with OAuth, with only the rights you grant it.

Why us

Four questions,
one answer.

  1. Do you keep my addresses?

    No.

    An address only passes through, for as long as it takes to verify it. All we keep is a one-way fingerprint, which cannot be turned back into the address, for the cache and aggregate statistics.

  2. Do you make up a verdict when you don't know?

    No.

    A provider that doesn't answer reliably gets a valid with a capped score and the reason spelled out. Anything we couldn't settle comes back not verifiable, never valid by default.

  3. Do I pay for a check that got no answer?

    No.

    If the mail server doesn't answer, the mailbox check is refunded. An address verified recently is served again for free, and every token shows up in your history.

  4. Do my lists leave the European Union?

    No.

    They are verified on our servers in Germany, with no processor outside the Union. Only each address's own mail server is asked, wherever it is, and only about that one address.

Your tools

Verify before you send, whatever tool you use.

Your Shopify signups, your HubSpot contacts, your Brevo campaigns: check every address before it goes out.

  • Gmail
  • Shopify
  • Brevo
  • Supabase
  • Mailchimp
  • Stripe
  • HubSpot
  • WordPress
  • Mailjet
  • Outlook
  • Google Forms
  • Resend
  • Webflow
  • lemlist
  • Firebase
  • Klaviyo
  • Calendly
  • Mailgun
  • Notion
  • Kit
  • Clerk
  • Loops
  • Wix
  • Intercom
  • Google Sheets
  • Campaign Monitor
  • Framer
  • Substack
  • Auth0
  • SparkPost
  • Squarespace
  • Zendesk
  • Airtable
  • Mailtrap
  • PrestaShop
  • Proton Mail
  • Gumroad
  • Ghost
  • Lemon Squeezy
  • HEY
  • Paddle
  • iCloud
  • BigCommerce
  • Thunderbird
  • PayPal
  • Odoo
  • ClickUp
  • Make
  • n8n
  • Discord
  • Claude
  • API
  • ChatGPT
  • Cursor
  • MCP
For your AI agents

Your agent verifies. You stay in control.

MCP is the standard that lets an AI assistant use an outside tool. Plug PricklyMails in once, and your assistant verifies an address or a whole list whenever you ask it to.

  • Same billing as the API: an agent running on its own only pays for the answers it gets, never for the attempts.
  • And it gets the same honest verdict: "not verifiable" rather than a guessed valid it would act on alone.
Read the MCP docs
  • Claude
  • ChatGPT
  • Cursor
  • Any MCP client
PricklyMails
Set up in three steps
  1. Add the server

    In your assistant's connector settings, paste this address.

    https://mcp.pricklymails.com/mcp
  2. Connect your account

    You sign in to PricklyMails and pick the agent that gets access, with its permissions. No key to copy, and you can revoke it whenever you like.

  3. Ask

    "Check these addresses and tell me which ones to remove." The agent picks the tool and hands you the result.

FAQ

Frequently asked questions

Verification

Every verdict comes from real checks: the address format, the domain and its mail servers, then, when possible, the receiving server itself, asked without sending any email.

Google, Microsoft and Yahoo don't answer that question reliably. For their addresses, you get a valid verdict with a capped score and the reason, rather than a confirmation they never gave.

When we can't tell, the verdict is "not verifiable". We never guess.

Integration

Data and GDPR

Plans and pricing

Check your list before your next send.

1,000 tokens free when you sign up, no credit card. After that, you only pay for answers.